Hpe Ase - Storage Solutions Architect V3 Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

HPE ASE—Storage Solutions Architect V3 OFFICIAL CERTIFICATION STUDY GUIDE (EXAM HPE0-J57) First Edition Radek Zima

HPE Press 660 4th Street, #802 San Francisco, CA 94107

||||||||||||||||||||

||||||||||||||||||||

HPE ASE—Storage Solutions Architect V3 Official Certification Study Guide (Exam HPE0-J57) Radek Zima © 2018 Hewlett Packard Enterprise Development LP. Published by: Hewlett Packard Enterprise Press 660 4th Street, #802 San Francisco, CA 94107 All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without written permission from the publisher, except for the inclusion of brief quotations in a review. ISBN: 978-1-942741-97-8 WARNING AND DISCLAIMER This book provides information about the topics covered in the Designing HPE Storage Solutions (HPE0-J57) certification exam. Every effort has been made to make this book as complete and as accurate as possible, but no warranty or fitness is implied. The information is provided on an “as is” basis. The author, and Hewlett Packard Enterprise Press, shall have neither liability nor responsibility to any person or entity with respect to any loss or damages arising from the information contained in this book or from the use of the discs or programs that may accompany it. The opinions expressed in this book belong to the author and are not necessarily those of Hewlett Packard Enterprise Press.

||||||||||||||||||||

||||||||||||||||||||

Feedback Information At HPE Press, our goal is to create in-depth reference books of the best quality and value. Each book is crafted with care and precision, undergoing rigorous development that involves the expertise of members from the professional technical community. Readers’ feedback is a continuation of the process. If you have any comments regarding how we could improve the quality of this book, or otherwise alter it to better suit your needs, you can contact us through email at [email protected]. Please make sure to include the book title and ISBN in your message. We appreciate your feedback. Publisher: Hewlett Packard Enterprise Press HPE Contributors: Wilfred Brown, Chris Hornauer, Ralph Luchs, Chris Smit HPE Press Program Manager: Michael Bishop

||||||||||||||||||||

||||||||||||||||||||

About the Author Radek Zima is an independent consultant who specializes in IT infrastructure design, implementation and maintenance for HPE servers, storage, networking, management and cloud software. He develops and delivers trainings, workshops, demonstrations, and conference presentations for HPE channel partners, customers and employees at training centers and events around the world. Radek has a Bachelor’s degree and a Master of Science degree from the Faculty of Informatics and Statistics, University of Economics in Prague.

Introduction This book is based on the Designing HPE Storage Solutions course and helps you prepare for the Designing HPE Storage Solutions (HPE0-J57) exam for the HPE ASE Storage Solutions Architect V3 certification. This certification validates that you can expertly explain, recommend, and design Enterprise Storage Solutions and translate business requirements into storage designs that support applications and data across physical, virtual and cloud environments with a common architecture and converged management. The certification also shows that you can design HPE Backup Solutions including the right Backup, Recovery, and Archive (BURA) strategies for various customer scenarios.

Certification and Learning Hewlett Packard Enterprise Partner Ready Certification and Learning provides end-toend continuous learning programs and professional certifications that can help you open doors and accelerate your career. We provide continuous learning activities and jobrole based learning plans to help you keep pace with the demands of the dynamic, fast paced IT industry; professional sales and technical training and certifications to give you the critical skills needed to design, manage and implement the most sought-after IT disciplines; and training to help you navigate and seize opportunities within the top IT transformation areas that enable business advantage today. As a Partner Ready Certification and Learning certified member, your skills, knowledge, and real-world experience are recognized and valued in the marketplace. To continue your professional and career growth, you have access to our large HPE community of world-class IT professionals, trend-makers and decision-makers. Share ideas, best practices, business insights, and challenges as you gain professional connections globally.

||||||||||||||||||||

||||||||||||||||||||

To learn more about HPE Partner Ready Certification and Learning certifications and continuous learning programs, please visit http://certification-learning.hpe.com

Audience This book is designed for presales architects and integrators or HPE channel partners who identify opportunities and define technical requirements for HPE Storage Solutions. It is recommended that candidates have a minimum of one to three years of experience in designing and implementing storage technologies.

Assumed Knowledge The HPE ASE — Storage Solutions Architect V3 certification Solutions certification is an intermediate-level certification. It is expected that candidates will have industrystandard storage knowledge from training, hands-on experience, or participation in other technical events.

Minimum Qualifications Pre-requisites to qualify for this certification are the HPE ATP - Storage Solutions V3 or the HPE ATP - Data Center Solutions V1.

Relevant Certifications After you pass these exams, your achievement may be applicable toward more than one certification. To determine which certifications can be credited with this achievement, log in to The Learning Center and view the certifications listed on the exam’s More Details tab. You might be on your way to achieving additional certifications.

Preparing for Exam HPE0-J57 This self-study guide does not guarantee that you will have all the knowledge you need to pass the exam. It is expected that you will also draw on real-world experience and would benefit from completing the hands-on lab activities provided in the instructor-led training. To pass the certification exam, you should…

||||||||||||||||||||

||||||||||||||||||||

Recommended HPE Training Recommended training to prepare for each exam is accessible from the exam’s page in The Learning Center. See the exam attachment, “Supporting courses,” to view and register for the courses.

Obtain Hands-on Experience You are not required to take the recommended, supported courses, and completion of training does not guarantee that you will pass the exams. Hewlett Packard Enterprise strongly recommends a combination of training, thorough review of courseware and additional study references, and sufficient on-the-job experience prior to taking an exam.

Exam Registration To register for an exam, go learning.hpe.com/tr/learn_more_about_exams.html

to

https://certification-

||||||||||||||||||||

||||||||||||||||||||

CONTENTS

1

HPE Storage Enterprise Market Prelearning check HPE Storage strategy for the Idea Economy HPE Storage strategy Data center is evolving rapidly Changes impacting customers Problems with legacy storage environments What do organizations need from their storage? Activity: Customer influences, drivers, and challenges Applications and data fuel the digital enterprise Infrastructure powering the Idea Economy Doing business in the Idea Economy Hybrid IT: The solution continuum The way infrastructure should work Deployment comparison Change the approach Optimizing the infrastructure for apps Storage types Learning check Storage trends Major market trends The HPE Vision HPE Solutions HPE Pointnext services HPE OneView Align storage goals with business goals How are IT and storage business demands being measured? Moving to Composable Infrastructure Why HPE? Gartner Magic Quadrant for solid-state arrays 2016 vs Gartner Magic Quadrant for general-purpose disk arrays Learning check Activity: HPE Solutions

||||||||||||||||||||

||||||||||||||||||||

Watch the OneView video at the link below: HPE Converged Infrastructure Key initiatives for long-term IT strategy HPE Converged Storage principles Converged Storage is designed to deliver ROI HPE Synergy—The first platform architected for composability Synergy composable storage Learning check Summary 2

HPE Storage Portfolio Prelearning check HPE Storage portfolio HPE Storage positioning HPE SimpliVity 380 Gen HPE SimpliVity backups Activity: Modernizing the infrastructure HPE StoreFabric Gen6 Fibre Channel HPE StoreEasy HPE StoreEasy 3850 WSS2016 Gateway Single Node Software-defined storage HPE StoreOnce portfolio StoreOnce single-node appliance StoreOnce single-node appliance—Key parameters StoreOnce 6600—Key parameters HPE StoreEver HPE T950 and TFinity ExaScale Edition Tape Library Learning check HPE primary storage HPE StoreVirtual HPE StoreVirtual Peer Copy from VSA to 3PAR VSA Ready Nodes Network RAID technology HPE StoreVirtual Peer Motion HPE SmartClone volumes

||||||||||||||||||||

||||||||||||||||||||

Learning check Activity: Using VSA Ready Nodes HPE 3PAR StoreServ HPE 3PAR StoreServ Family HPE 3PAR StoreServ 9450 and 20000 provide highest resiliency HPE 3PAR StoreServ 9450 performance HPE 3PAR StoreServ HPE 3PAR StoreServ—Converged flash 8440 and all-flash HPE 3PAR StoreServ HPE 3PAR StoreServ architecture HPE 3PAR ASIC Features of StoreServ architecture No single point of failure with StoreServ components Converged block, file, and object access HPE 3PAR StoreServ: Embracing new technologies Architected for efficiency today and for what comes next HPE 3PAR gets you ready for what is next Fifty percent lower average latencies and 80% higher IOPS HPE InfoSight for 3PAR Portal navigation Learning check Activity: Understanding 3D Cache HPE Nimble Storage Nimble All-Flash storage family Scale out with clustered arrays Nimble Storage graphical user interface (GUI) HPE InfoSight HPE InfoSight Predictive Analytics Capacity trending and upgrade recommendations Benefits of radical simplicity are clear and measurable Secondary Flash Arrays—Put your backup data to work Simplified data management Twenty percent more usable capacity than other vendors Viewing data reduction savings Triple parity protection plus spare drive Radically simple “Timeless Storage” business model HPE Cloud Volumes

||||||||||||||||||||

||||||||||||||||||||

Data protection Multicloud without lock in Positioning Nimble within the HPE Storage portfolio Activity: Artificial intelligence for the data center HPE management tools HPE 3PAR StoreServ Storage management Federation view in SSMC HPE 3PAR StoreServ CLI HPE Smart SAN HPE Insight Remote Support (IRS) HPE Storage Operations Manager HPE Recovery Manager Central HPE OneView—Deploy infrastructure faster HPE OneView licensing Learning check HPE 3PAR StoreServ Operating System and Software HPE 3PAR StoreServ Software portfolio HPE 3PAR 7/10000 StoreServ Software Suites 3PAR 8000 and 20000 software suites 3PAR 8000, 9000, and 20000 software details All-inclusive licensing for existing 8k and 20k customers HPE 3PAR Adaptive Data Reduction HPE Express Layout HPE 3PAR Thin Technologies HPE 3PAR Thin Deduplication HPE 3PAR Data-at-Rest Encryption Workload-centric storage personas HPE 3PAR StoreServ File Persona HPE 3PAR File Persona HPE 3PAR File Persona logical architecture HPE 3PAR File Persona features HPE 3PAR Dynamic Optimization HPE 3PAR Adaptive Optimization HPE 3PAR Priority Optimization (QoS) HPE 3PAR Adaptive Flash Cache Adaptive Flash Cache compared to Adaptive Optimization HPE 3PAR StoreServ Peer Motion and Online Import

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR Persistent technologies HPE 3PAR Persistent Cache Persistent Ports Persistent Ports on HPE 3PAR StoreServ systems HPE 3PAR Remote Copy software Asynchronous Periodic mode replication Asynchronous Streaming mode replication HPE 3PAR Storage Federation Learning check Summary 3

HPE Virtualization, Fabrics, and Converged Management Prelearning check Fundamentals of virtualization Journey of virtualization Storage integration into hypervisors Software-defined storage HPE StoreVirtual VSA HPE StoreVirtual VSA Ready Nodes Learning check VMware Virtual Volumes VVol integration with shared storage Desktop virtualization HPE Nimble solutions for virtual desktop infrastructure HPE InfoSight VMVision VMVision Datastore TreeMap Latency analysis Virtual machine view Operational Dashboard VMware Horizon conceptual architecture Learning check Activity: Implementing VMware VVs on HPE 3PAR StoreServ HPE Virtual Connect FlexFabric FlexFabric and HPE 3PAR StoreServ HPE flat SAN architecture Convergence with Virtual Connect FlexFabric modulesand adapters

||||||||||||||||||||

||||||||||||||||||||

Virtual Connect FlexFabric adapters Learning check SAN fabric topologies HPE SAN support Levels of availability Single-switch fabric Meshed-switch fabric Multiple fabrics and device paths Routed SAN fabrics Learning check Activity: Using the SAN Design Reference Guide Virtualization management with HPE OneView HPE OneView Consumer-inspired user experience HPE OneView for storage management The best of control and agility with storage templates StoreVirtual automation—HPE OneView Management platform built on the unified REST API HPE OneView: Your infrastructure automation engine HPE OneView licensing Managing storage resources with HPE OneView HPE OneView and HPE 3PAR StoreServ systems Deploying applications using rapid automation and growth Learning check Activity: HPE OneView Licensing Guide video HPE Converged Systems HPE ConvergedSystem Hyperconvergence—Support for a hybrid infrastructure What is HPE Hyper Converged? The evolution of hyperconvergence HPE SimpliVity HPE Composable Infrastructure Benefits of Composable Infrastructure The versatility of Composable Storage via HPE OneView HPE Synergy Activity: Using HPE Reference Architecture Learning check

||||||||||||||||||||

||||||||||||||||||||

Summary 4

HPE Storage Services Prelearning check Partner Ready Services Value in partnership Business benefits of the Partner Ready Services Earning competencies Activity: Accessing the Partner Ready Portal Learning check Overview of HPE Storage support services Support Services Central HPE Support Center access Firmware updates for eligible products Auditing and call-to-repair times Activity: Understanding severity levels: Part Activity: Understanding severity levels: Part Packaged service levels HPE Foundation Care HPE Foundation Care support levels HPE Proactive Care HPE Proactive Care Advanced HPE Lifecycle Event Services HPE Insight Remote Support HPE Datacenter Care Activity: Proactive Care Advanced video Learning check HPE Financial Services HPE Financial Services adding value for channel partners Our promise to our partners HPE Renew program Activity: Understanding the Financial Services sales steps HPE Consulting Services Simplify IT with HPE Pointnext Services HPE Services portfolio Professional Services Operational Support Services

||||||||||||||||||||

||||||||||||||||||||

Data Storage Services HPE Solution Consulting Services HPE 3PAR StoreServ Data Migration Service HPE Network Transformation Experience Workshop Learning check Summary 5

Planning HPE Storage Solutions Prelearning check Choosing the most appropriate drive type Hard disk drives IO workloads Considerations while choosing HDDs HDD types Solid-state drives Choosing an HDD or SSD solution for an application Comparing drive performance Drive IO and RAID performance impact Adaptive Flash Cache Using AFC to increase read hits and improve randomread performance Moving data from a node’s DRAM read cache into AFC HPE SmartCache architecture SmartCache—DAS Hybrid flash rules and guidelines for Nimble Learning check Activity: Choosing the correct SSD Planning a RAID configuration 3PAR StoreServ RAID 6 concept Using network RAID in a StoreVirtual Cluster Planning and designing a SAN HPE approach to SAN design SAN design principles and considerations SAN components Topology data access performance Fibre Channel and converged switches HPE Flex-10 or FlexFabric design HPE Virtual Connect for Synergy

||||||||||||||||||||

||||||||||||||||||||

HPE Synergy composable fabric HPE StoreFabric M-Series Ethernet switches—Optimizedfor storage HPE FlexFabric solution product portfolio HPE FlexFabric switches Learning check Activity: Understanding SAN extensions Configuring the most resilient HPE 3PAR configuration NinjaSTARS: Elements to be considered Adaptive Data Reduction HPE 3PAR StoreServ 9450 hardware building blocks HPE 3PAR StoreServ 9450 controller HPE 3PAR StoreServ 9000 configuration considerations Choose drive and host adapter HPE 3PAR StoreServ 9450 port scalability Configuring the 8000, 9450, and 20000 for utmostperformance Choose cables Choose Service Processor HPE 3PAR StoreServ 9450 racking options HPE 3PAR StoreServ 9450 warranty, support,and installation HPE 3PAR StoreServ 9450 performance Learning check Activity: Using NinjaSTARS to create a configuration Examples from NinjaSTARS Activity: Using Nimble Config Sizer to create a configuration HPE Reference Architectures HPE Converged Architecture 700 with Arista Activity: Reference Architecture Learning check Summary 6

Staging an Effective Storage Consulting Engagement Prelearning check Presales consulting goals Customer culture Recognize the buyer, their top priority, and desired outcomes Align the right solution by workload and use case Use the flash conversation to identify dataprotection opportunities

||||||||||||||||||||

||||||||||||||||||||

Do not miss the secondary storage opportunity Consulting engagement stages Where do projects come from? Learning check Business value sales with storage Engage early in the sales cycle with consultative selling HPE Storage business value methodology Defining business value in storage Addressing business objectives Steps to a successfully architected storage solution Four fundamental storage architecture views HPE sizing and planning tools HPE Storage Sizing Tool HPE Nimble Config Sizer Storage Assessment Foundry Storage Assessment Foundry Suite HPE NinjaSTARS for HPE 3PAR StoreServ How to achieve better data protection Storage Optimizer One Config Advanced Learning check Activity: Completing the OCA introduction Using TCO and ROI tools TCO ROI Alinean ROI and TCO analysis HPE Storage Quick ROI Calculator Alinean customer deliverables Technical Enablement Tools and Resources Resources and tools HPE Demonstration Portal HPE Storage Information Library HPE QuickSpecs HPE SAN certification and support Single Point of Connectivity Knowledge Learning check Activity: Using the SAN Design Reference Guide

||||||||||||||||||||

||||||||||||||||||||

Proof of Concept Test guidelines HPE PoC App VSA design considerations Learning check Summary 7

Administering and Monitoring HPE Storage Solutions Prelearning check HPE 3PAR manageability 3PAR—Management options HPE 3PAR StoreServ administration Using the SSMC Learning check Activity: HPE Smart SAN for 3PAR video HPE 3PAR StoreServ System Reporter Threshold alerts Excel add-in for SSMC HPE 3PAR Service Processor Service Console VSP infrastructure planning and deployment best practices Remote support Learning check Activity: Prerequisite for setting up the Service Processor Managing HPE StoreOnce systems Reporting Central—Reports Firmware updates Configuration save Confidence Checker HPE StoreOnce RMC for VMware software (RMC-V) Installing Recovery Manager Central RMC—Fibre Channel connectivity Learning check Nimble Storage system NimbleOS home page Administration menu How does InfoSight help customers?

||||||||||||||||||||

||||||||||||||||||||

Identifying where the problems exist Learning check HPE StoreVirtual storage systems Dashboard view Centralized Management Console main window HPE OneView HPE OneView storage management HPE OneView for VMware vCenter HPE OneView for Microsoft System Center Learning check Activity: What is new with OneView 4.0? Learning check Summary 8

Upgrading, Optimizing, and Tuning HPE Storage Solutions Prelearning check Upgrading the HPE 3PAR StoreServ system High availability Adding new components to an existing system Controller node upgrade Drive enclosure upgrade Disk chassis availability Drive and enclosure upgrades Allocating and loading guidelines—SFF drives Allocating and loading guidelines—LFF drives Allocating and loading guidelines—Avoid practices Disk drive upgrade Learning check Activity: HPE 3PAR StoreServ 9000 Storage CustomerSelf-Install video HPE InfoSight and upgrading HPE Nimble systems Upgrading the HPE Nimble system What is in a CSx000 Array? Drive layout—CS1000, CS3000, CS5000, and CS Drive layout—DFC details Drive layout—CS1000H: A half-populated CS CS-Series connectivity options Nimble Storage CSx000 scale to fit

||||||||||||||||||||

||||||||||||||||||||

CSx000 controller upgrades CS1000H “half-pop” controller upgrades ES2 Expansion Shelf Drive layout—ES Upgrading HPE StoreOnce Systems HPE 5100 Capacity Upgrade system components Firmware upgrades Upgrading firmware Multinode systems Download Quick Restore ISO Image Learning check Optimizing HPE StoreOnce Storage systems Optimizing StoreOnce Catalyst best practices Optimization considerations with deduplication on StoreOnce Optimizing general best practices: HPE 3PAR StoreServ 8000initial configuration HA disk: Two-node drive upgrade best practice HA disk: Two-node enclosure upgrade best practice 3PAR OS 3.3.1 and active-active disks How set size affects rebalancing activities and upgrades Mixed-drive capacities and speeds All-flash array considerations Adaptive Flash Cache Increasing and decreasing the size of Adaptive Flash Cache HPE 3PAR Dynamic Optimization HPE 3PAR Adaptive Optimization HPE 3PAR Priority Optimization Learning check Tuning HPE 3PAR StoreServ systems Introduction to tunesys Tunesys phases Tunesys SSMC—Accessing Tuning from SSMC Learning check HPE 3PAR StoreServ operating system upgrade Planning information for online HPE 3PAR OS upgrades HPE 3PAR StoreServ Storage configuration guidelines Preinstallation requirements

||||||||||||||||||||

||||||||||||||||||||

Starting HPE 3PAR OS software update from the SP Update HPE 3PAR OS HPE 3PAR StoreServ Service Processor upgrade Uploading an ISO image Service Processor Update page Activity: HPE 3PAR StoreServ 7000 Storage Remove/Replacedisk video Learning check Summary 9

HPE Backup Solutions Prelearning check HPE StoreOnce overview Block, file, and object compared HPE StoreOnce portfolio StoreOnce Deployment planning and sizing considerations Choosing the most appropriate StoreOnce system Configuration best practices—General Performance: Understand performance curve progression Performance considerations Additional considerations StoreOnce VSA Backup target emulation types—VTL Backup target emulation types—NAS Learning check StoreOnce Catalyst HPE StoreOnce Catalyst ISV integration Multihop HPE StoreOnce deduplication Federated deduplication Federated StoreOnce Catalyst StoreOnce Catalyst principles STaTS support for customers with 3PARand StoreOnce StoreVista individual system detail and performance history Nimble SF-Series Secondary Flash Array Veeam integration Learning check Activity: StoreOnce Systems to deal with data growth video

||||||||||||||||||||

||||||||||||||||||||

Recover Manager Central RMC components Monitoring RMC Appliance—Dashboard Peer Copy Express Protect Express Restore Element Recovery Telemetry Guided workflow—new way to protect the data RMC 4.1 has the ability to suspend schedules RMC-V Snapshots/Veeam Explorer Integration Learning check Activity: Creating snapshots and Express Protect backupfor Oracle databases video StoreOnce operating system features and benefits Hardware Appliances: StoreOnce Virtual Appliances: HPE StoreOnce Cloud Bank Storage Reporting Central Catalyst Copy over Fibre Channel Learning check Activity: HPE Cloud Bank Storage video HPE StoreEver Tape portfolio HPE StoreEver—LTO Tape Drive/MSL Tape Libraries HPE LTO Tape Drive portfolio LTO technical specifications HPE StoreEver 1/8 G2 Tape Autoloader family comparison HPE StoreEver MSL 6480 comparison HPE TFinity ExaScale Edition BlueScale tape library software New HPE T950V with LTOLearning check HPE tape tools StoreEver TapeAssure Advanced Software and CV-TL HPE Library and Tape Tools Ease of use with HPE StoreOpen and LTFS What is NinjaProtected+? Storage Assessment Foundry HPE Storage Sizer input files

||||||||||||||||||||

||||||||||||||||||||

HPE Switch Selector tool Learning check Activity: Hewlett Packard Enterprise Switch Selector Learning check Summary 10 Designing HPE BURA Solutions Prelearning check Availability and downtime Planned downtime Activity: Identifying causes of downtime What is backup? What is restore? Identifying the data to be saved Differential backup Incremental backup When to choose incremental vs. differential backup Synthetic backup Conceptual illustration of RMC Express Protect—Inlinesynthetic full Archive bit RPO, RTO, and retention Differences between a backup and archiving Tiered data retention for HPE StoreServ Learning check Activity: Calculating the cost of downtime Backup strategy planning The data protection continuum Defining the requirements of a backup and recovery strategy Preparing the backup design The infrastructure methods and devices Capacity and sizing Capacity planning usage models Learning check Array snapshot with HPE RMC-V Impact of traditional backup on mission-criticalapplications in all-flash 3PAR environments Impact of an incremental backup on mission-criticalapplications in all-flash

||||||||||||||||||||

||||||||||||||||||||

3PAR environments Eliminating backup impact on mission-critical applicationsin all-flash 3PAR environments with 3PAR RMC Snapshots alone cannot deliver comprehensive applicationprotection in allflash environments Protecting workloads at the speed of 3PAR Flash Flash-integrated data protection without compromise Sizing an RMC solution Learning check Software to support HPE BURA Traditional backup and archive is unsustainable HPE BURA Data Protector Data Protector 10. Data Protector Architecture How deduplication integrates into Data Protector Target-side deduplication with Data Protector and StoreOnce Server-side deduplication with Data Protector and StoreOnce Source-side deduplication with Data Protector and StoreOnce Capacity planning considerations Activity: Sample backup information collection forcapacity planning Micro Focus Adaptive Backup and Recovery Recovery Manager Central and RMC-V Veeam and StoreOnce integration Comparison of several item recovery options withinVeeam in HPE Storage environment Integrate 3PAR Integration with Veeam Veeam with Nimble highlights Backup from Storage snapshots Secondary Flash Array Use Cases Storage Optimizer Learning check Compliant archiving challenge iTernity Compliant Archive Software HPE Complete—iCAS HPE StoreEasy Storage and iCAS Addressing emerging use cases—Compliance and security

||||||||||||||||||||

||||||||||||||||||||

HPE Enterprise Secure Key Manager StoreOnce Security Pack Local key and centralized key management Learning check Summary Practice Test Introduction Ideal candidate for this exam HPE0-J57 testing objectives Test preparation questions and answers Questions Practice Test Answers and explanations Learning Check Answers Chapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter 5 Chapter 6 Chapter 7 Chapter 8 Chapter 9 Chapter 10

||||||||||||||||||||

||||||||||||||||||||

1

HPE Storage Enterprise Market

||||||||||||||||||||

||||||||||||||||||||

LEARNING OBJECTIVES After completing this chapter, you should be able to: Describe the Hewlett Packard Enterprise (HPE) Storage strategy in the Idea Economy Summarize Storage trends Explain HPE Solutions Identify HPE Converged and Composable Infrastructure

||||||||||||||||||||

||||||||||||||||||||

Prelearning check To survive in today’s fast-paced enterprise environment, businesses need to become more agile. What are the three key components fundamental to HPE Composable Infrastructure that enable agility?

||||||||||||||||||||

||||||||||||||||||||

HPE Storage strategy for the Idea Economy The increasing complexity of the storage decision reflects the growing recognition of storage as a strategic investment. The storage strategy therefore needs to be better aligned to support customers’ applications and business goals. HPE Storage strategy encompasses System Defined and traditional storage as well as Software Defined and Composable Infrastructure, as shown in Figure 1-1.

Figure 1-1 System- and software-defined storage HPE believes that you will probably need both, especially if you want to build a Data Center Architecture for the New Style of IT. Today more than ever before businesses need to find the right strategy for their IT. Although they scramble to transform themselves to compete in the digital economy, pressure is placed on them to get it right. According to Frost & Sullivan research, 40% of CEOs globally are focusing on their technology strategy to drive revenue growth over the next three years. Among IT decision makers, 73% are pursuing a hybrid IT strategy.

||||||||||||||||||||

||||||||||||||||||||

HPE Storage strategy Storage administrators need to address many different challenges: What storage technology (flash or HDD)? What storage structure (block, file, object)? What infrastructure (physical or virtual, standalone or converged)? What deployment model (premises-based, cloud, or hybrid)? As shown in Figure 1-2, more and more businesses today are moving toward a hybrid cloud environment for their IT.

Figure 1-2 Primary deployment model used by businesses According to a Frost & Sullivan survey of IT decision makers: 73% of businesses currently have or plan to implement a hybrid cloud by next year. 71% of businesses expect that their hybrid cloud will include their own data center as well as various vendors’ public and private cloud services. Regardless of how comprehensive your strategy is or how committed your company maybe toward transforming to a hybrid IT environment, few companies implement a

Technet24

||||||||||||||||||||

||||||||||||||||||||

full-scale infrastructure upgrade. Such an initiative is usually deemed too costly, in terms of time, budget, disruption, and business risk. It is more likely that you will select a few storage optimization projects to get started. Cloud and hybrid infrastructures are not just the preserve of enterprise customers. Most SMBs (small and medium-sized businesses) understand that they need to put technology to work to transform their businesses for the future. Of total, 72% of SMB decision makers say that technology solutions can help them significantly improve business outcomes and/or run the business better, and 53% plan to increase technology investments. Although most SMBs do not have the internal IT resources required to take advantage of technology innovation, the cloud has become the backbone to enable this. SMBs are increasingly opting to put new workloads in the cloud, relying on cloud providers to supply, manage, and maintain IT infrastructure and applications for them. SMB adoption of cloud-based business solutions continues to rise steadily across all business solution categories.

Data center is evolving rapidly As shown in Figure 1-3, data centers have evolved from onsite mainframe mega machines to private, public, and hybrid clouds. A decade ago, an infrastructure could be deployed and consistently managed for about five years. In recent years, that planning horizon has decreased to one or two years. Megatrends that are shaping the next generation of data centers include: Mobile and ubiquitous computing—Increasingly digital capabilities allow multidevice access from any location. More software-as-a-service (SaaS)—As SaaS strategies increase, data-centers-as-aservice will increase too. A majority of an enterprise could become a service, allowing a company to focus more on its core competencies. Increasingly nimble competitors—New entrants with low barriers to entry and a distinct lack of legacy systems will challenge entrenched companies. Role convergence—Business leaders will need to understand IT and vice versa. Time, effort, and resources are needed to translate business-user requirements into IT specs. If the translation steps were eliminated, imagine how much faster you could develop systems. Workplace transformation—Employees can work from anywhere and collaborate with

||||||||||||||||||||

||||||||||||||||||||

anyone in real time. Companies can tap talent from anywhere and lose talent to anyone.

Figure 1-3 The journey to cloud

Changes impacting customers Customers say that legacy storage is too complicated to manage, expensive, hard to scale, disconnected, inefficient, and inflexible. Unpredictable workloads and explosive data growth expose performance and scalability limitations in legacy architecture. It is no longer realistic to tune an entire array around a single type of data or single application. Taking storage offline or disrupting applications simply to add new capacity is not an option. Enterprises realize that their current infrastructure is not adequate for future requirements, as shown in figure 1-4. Industry innovations and customer demand are driving change.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 1-4 Customer challenges Market trends affecting the storage infrastructure include: Big data analytics tools will be used on data collected from the Internet of Things (IoT) to fundamentally change how IT interacts with the world. As storage architectures incorporate flash drives, companies are reducing the price they spend per gigabyte/terabyte by using deduplication and thin technologies. As customers move from near-term private clouds to longer-term public clouds, software-defined storage (SDS) will bridge private and public clouds. Hyperconverged systems comprise a fast-growing part of the overall Converged Infrastructure market. IT is moving from a project-based model to continuous scaling. Requirements are being fulfilled from the resource pool. Implementing new solutions rapidly is critical to staying competitive. Data protection, replication, and management software will need to be tightly integrated (and included)

||||||||||||||||||||

||||||||||||||||||||

within a customer offering.

Problems with legacy storage environments IT environments with a large number of different storage systems and architectures are complex, costly, and rigid. These types of environments cannot meet customer needs for primary storage. As shown in figure 1-5, traditional systems can be: Too complex and slow to deploy—Too many vendors and architected silos can restrict agility by breaking processes into several different administrative handoff tasks. Often, IT decisions do not account for long-term business needs when an architected, systematic approach is more appropriate. Too expensive and inefficient—Massive underutilizations can waste up to 70% of a company’s budget. Unfocused attempts to fix the problem can lead to overspending on capital, power, cooling, and administration. Too rigid with an inflexible infrastructure—Static systems cannot meet unpredictable service level, compliance, or Big Data requirements. They are designed for traditional, established platforms and predictable environments.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 1-5 Problems with legacy storage As an example, in the banking sector, a company might have three different tiers and classes of storage, including Tier 1 for mission-critical applications, a midrange platform for general business applications, and solid-state drives (SSDs) for highfrequency trading applications. The company might use only one leading vendor with three completely disconnected storage platforms. Typically, these systems are not managed the same and cannot exchange data. Disconnected disparate silos present issues as well. For unstructured data, another pattern often emerges. A general-purpose file-tier structure might scale out to help consolidate. The company should consider object storage to help share large files and avoid having internal developers move data into the public cloud.

What do organizations need from their storage? Storage solutions must be aligned with business strategies and objectives. As shown in figure 1-6, customer demands and expectations include: Seamless delivery of IT services—This demand is partly driven by the consumerization of IT. This translates to primary storage for applications independent of data type (object, file, or block) or delivery mechanism (physical, virtual, or cloud). User expectations are higher because of the instantaneous access to applications and data that they experience with smartphones, tablets, and so on. More value from information—To get more value from data and information, customers focus on Big Data. Companies are keeping data for longer time periods and need to manage that lifecycle accordingly. Performing analytics queries and using meaningbased computing models on archived data can be difficult and inconvenient, but it is necessary to make better informed decisions moving forward. Reduced risk exposure—As businesses become increasingly dependent on IT, the cost of downtime is at an all-time high. Storage environments focus not only on changing the economics of risk reduction but also on ensuring that data can be recovered quickly when needed.

||||||||||||||||||||

||||||||||||||||||||

Figure 1-6 What do organizations need?

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Customer influences, drivers, and challenges Instructions: As we go through this course, we will explore some of the storage solutions and tools that HPE has to offer and get a better understanding of how we can align these to meet our customers’ challenges. Even if your storage today is confined to a single environment, knowing that you might eventually require a hybrid environment will enable you to start investing in platforms designed to support hybrid solutions. Where are your customers along this journey and what has motivated them? Influences: Reasons for change

Drivers: Market and business forces

Challenges: Issues: Current and future

Thinking about your customers, note your thoughts under each of the above headings.

Applications and data fuel the digital enterprise We are living in an Idea Economy, where the ability to turn an idea into a new product or service has never been easier. In today’s hyperconnected world, applications are the mechanism for delivering innovative ideas, and we use apps for everything from our day-to-day life to business. Apps make life easy, but they also make it easy to switch suppliers in an instant, so maintaining customer loyalty requires providing a better customer experience through continuously delivering new app features. Every business is now a digital business, and apps are the digital products driving innovation and causing market disruption.

||||||||||||||||||||

||||||||||||||||||||

Infrastructure powering the Idea Economy The evolution from traditional, siloed infrastructure to Composable Infrastructure is shown in figure 1-7. Traditional infrastructure is often siloed and burdened with complex processes. It can take IT months to stand up environments for traditional applications—regardless of whether these environments are meant to support physical, virtual, or containerized workloads.

Figure 1-7 From traditional to Composable Infrastructure Converged Infrastructure with hyperconverged solutions is extremely agile and only requires minutes to deploy. These workload-specific offerings deliver fluid virtual IT and SDS in a single appliance. HPE Converged Infrastructure and hyperconverged appliances help organizations succeed in the Idea Economy. Doing business in the Idea Economy—which is also called the digital, application, or mobile economy—means turning an idea into a new product, capability, business, or industry. To stay competitive, organizations need to develop new IT capabilities, create new outcomes, proactively manage risk, be predictive, and create a hyperconnected workplace. HPE Converged Infrastructure, software-defined management, and hyperconverged systems reduce costs and increase operational efficiency by eliminating silos and freeing available compute, storage, and networking resources. However, most organizations must still maintain their traditional infrastructure, which can involve using a different set of tools and APIs (application programming interface — communication interface between software components). Businesses need a single, fluid infrastructure that is optimized to deliver infrastructure for both traditional and Technet24

||||||||||||||||||||

||||||||||||||||||||

new applications in seconds. This infrastructure must be agile and flexible enough to change personalities dynamically so that it can meet traditional and idea-economy workload demands without overprovisioning resources for either. This architectural approach is a Composable Infrastructure.

Doing business in the Idea Economy Agility is key to surviving in the Idea Economy where things change at break-neck speed, you need to be able to adapt very quickly. For an enterprise solution to be able to compete in such a fast-paced market place, we need infrastructure that is capable of keeping up with these changes. It is important to focus on simplifying and automating the server, storage, and networking infrastructure. Business is changing; gone are the days when implementation of new business took months even years. In today’s Idea Economy, we need to be agile enough to respond to change in weeks, even days to stay ahead of the competition, and Composable Infrastructure helps us to do this. As shown in figure 1-8, three key components are required to orchestrate these rapid changes: Fluid resource pools meet each application’s changing needs by allowing for the composition and recomposition of single blocks of disaggregated compute, storage, and fabric resources needed to run applications in sizes at the appropriate scale. This allows new applications to be quickly provisioned. Software-defined intelligence provides a single management interface to integrate operational silos and eliminate complexity. Workload templates speed deployment, and frictionless change eliminates unnecessary downtime. The unified API provides a single interface to discover, search, inventory, configure, provision, update, and diagnose the Composable Infrastructure. A single line of code enables full infrastructure programmability and can provision the infrastructure required for an application.

||||||||||||||||||||

||||||||||||||||||||

Figure 1-8 Key components of Composable Infrastructure

Hybrid IT: The solution continuum Digital enterprises must manage apps and services using a single as-a-service IT model and optimize workload placement across hybrid IT to balance workload service-level agreement (SLA) needs. Digital transformation requires: Standardization, consolidation, and virtualization across the data center to reduce cost and improve utilization Automation to orchestrate IT processes and tasks to improve operational efficiencies Transformation to a DevOps delivery model to accelerate application and service delivery for business agility The HPE portfolio of solutions and services enables organizations to meet this Technet24

||||||||||||||||||||

||||||||||||||||||||

challenge and deliver business outcomes. HPE makes hybrid IT simple!

The way infrastructure should work With the emergence of cloud and other technologies, customers need faster operations to drive more profitability. As shown in figure 1-9, HPE customers want to be relevant to and succeed on both ends of the application spectrum. IT needs a more efficient approach that can create and deliver new value instantly and continuously.

Figure 1-9 Two operational models of IT New platforms need to: Run anything—Optimize any application and store all data on a single infrastructure with fluid pools of physical and virtual compute, storage, and fabric. Move faster—Accelerate application and service delivery through a single interface that precisely composes logical infrastructures at near-instant speeds. Work efficiently—Reduce operational effort and cost through internal software-defined intelligence with template-driven, frictionless operations. Unlock value—Increase productivity and control across the data center by integrating and automating infrastructure operations and applications through a unified API. With HPE Synergy, customers can: Dramatically reduce overprovisioning, capital expenditure (CapEx), and stranded resources

||||||||||||||||||||

–

Optimize SDS, direct-attached storage (DAS), and storage area network (SAN)

–

Compose any compute with any storage pool (including non-SAN simplicity)

–

Quickly recompose within a frame, across racks and rows in seconds

Deploy at cloud-like speed –

Stand up the infrastructure in minutes, not months

–

Auto-assemble all resources

–

Quickly deploy the configuration, drivers, firmware, and operating system through workload-based templates (catalog of images)

Develop apps and program infrastructure based on their preferences –

Program all resources using a single line of code

–

Program infrastructure at near run time

–

Store images for fast iteration and repeatable scale

||||||||||||||||||||

Note An operating expense, operating expenditure, operational expense, operational expenditure, or OpEx is an ongoing cost for running a product, business, or system. Its counterpart, a CapEx, is the cost of developing or providing nonconsumable parts for the product or the system. For example, the purchase of a photocopier involves CapEx, and the annual paper, toner, power, and maintenance costs represent OpEx. For larger systems like businesses, OpEx might also include the cost of workers and facility expenses such as rent and utilities (Wikipedia.org, January 2018).

Deployment comparison Figure 1-10 compares the right-sized Composable Infrastructure model to that of traditional enterprise environments and cloud service providers.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 1-10 Infrastructure types compared Cloud service providers can be two orders of magnitude larger than traditional enterprise data centers and often must scale to millions of servers. They have the luxury of buying thousands of servers and the associated networking and storage at one time at deep discounts. They provision everything in advance to meet the specifications of a handful of unique applications. Then, they simply turn on pre-configured, pre-defined, and pre-wired equipment and provisioning for the application they want to run. By comparison, the traditional enterprise hardware implementation process usually takes much longer because a majority of their applications carry unique requirements. Traditional data centers cannot match the efficiency of the cloud service providers’ custom server environments while maintaining traditional infrastructure. Unlike cloud service providers, enterprise IT cannot choose the infrastructure it needs until the applications are selected. Also, server, network, and storage equipment are often purchased by separate IT teams and then brought together for provisioning—a process that can take weeks or even months to coordinate and complete.

Change the approach New enterprises can easily implement the right-size infrastructure to run their applications. This enables transformational capabilities: Compute, storage, and fabric are integrated. Flexible and composable hardware runs all applications. Single pane of glass for automated operation and maintenance.

||||||||||||||||||||

||||||||||||||||||||

If customers want to gain the efficiencies of cloud service providers, they need to adopt their approaches, strategies, and business tactics. The Synergy Composable Infrastructure integrates compute, storage, and fabric in one solution, so enterprises can easily implement the right-size infrastructure to run their applications. This enables transformational capabilities: Compute, storage, and fabric are brought together—Pre-configured, more flexible, and ready to run, a Composable Infrastructure can be provisioned on demand as applications require. This dramatically cuts the time needed for setup and deployment of new infrastructure to run a new application or IT service. IT can optimize operations and meet business demands while still effectively managing efficiency and scale. Flexible and composable hardware runs all applications—Synergy has both resiliency and flexibility at the hardware level, so it can adapt to the needs of virtually any application. It has the unique ability to run without being limited to any one computing paradigm because it can run virtual machines, bare-metal deployments, containers, and cloud-native applications. Synergy allocates fluid pools of resources, which allows the Composable Infrastructure to run and store virtually anything. Synergy can disaggregate and reaggregate underlying blocks of resources with physical, virtual, and container resource pools. To meet application requirements, storage can be directly attached, network-attached, or software-defined and be configured to present data services in block, file, or object storage formats. This provides the flexibility to respond effortlessly to unpredictable demands, regardless of data type, connectivity protocol, or service level requirement.

Optimizing the infrastructure for apps Data centers are in a period of transition. The trend that started over a decade ago—to drive pervasive virtualization with x86 servers—shows no signs of slowing. Virtualization turns physical devices into a set of resource pools that are independent of the physical asset they run on. IT organizations have realized that by decoupling the application from hardware, workload mobility and higher utilization of shared physical server resources are possible. With over 70% of production systems virtualized, IT organizations are now focused on boosting utilization by increasing VM-to-server density (virtual machine to hypervisor). Yet realizing the true benefits of server, virtualization is thwarted when the underlying data center infrastructure is not virtualized. The mobility of VMs is limited when a datastore in the virtualization domain is tied to a physical storage system and its management is limited when it is

Technet24

||||||||||||||||||||

||||||||||||||||||||

dictated by a monolithic storage construct such as a LUN. Achieving virtualization across every element of the IT environment, with the goal of automated management and improved service delivery via software is the next step toward a software-defined data center. Unfortunately, legacy infrastructure stacks do not provide this level of data virtualization. The resulting environment is complex and costly to manage, and the scale only exacerbates the problem. It should come as no surprise that legacy architectures inhibit virtualization and today’s modern data mobility needs, especially since these architectures were optimized for a different world—a world that predated server virtualization, cloud, and SSDs. It was the world in which IT was tightly centralized within a single data center, and if remote IT sites existed, they were run independently from the primary data center. The size of disk drives was relatively small, the cost of storage was very high, replication was a luxury—very few could afford, and most disaster recovery operations were enabled by tape media. The situation resulted in operational inefficiency—more infrastructure to manage, reactive “firefighting” activities, and baseline service level delivery. Yet while underlying data architectures have changed very little, IT’s role in business has changed dramatically. IT—and the data it is chartered to protect—is at the core of almost all business today. As such, the demand on IT is ever increasing. To address the demand, IT has been forced to deploy a variety of technologies to solve disparate challenges in the data center. The last 10 years have seen a wave of specialty devices— such as wide area network (WAN) optimization, disk-based backup, SSD storage acceleration devices, and cloud gateways—each independently delivering value but collectively adding to the complexity crisis.

Storage types In today’s storage industry, different types of storage will be required for different types of data; there is no “One size fits all”. Getting the most efficient storage type for your data is an important as getting the right storage. Figure 1-11 sourced from https://insights.ubuntu.com/2015/05/18/what-are-thedifferent-types-of-storage-block-object-and-file/.

||||||||||||||||||||

||||||||||||||||||||

Figure 1-11 File, block, and object storage

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check To survive in today’s fast-paced enterprise environment, businesses need to become more agile. What are the 3 key components fundamental to HPE Composable Infrastructure that enable agility?

Storage trends The data center is undergoing its most dramatic shift since the move to client/server computing 20 years ago. The changes impacting customers include explosive growth, new workloads, virtualization, automation, cloud computing, and infrastructure and technology shifts such as in-memory databases. As shown in figure 1-12, there are two major storage trends—All Flash systems and SDS.

Figure 1-12 Storage trends

||||||||||||||||||||

||||||||||||||||||||

Both have different deployment models: All Flash optimized Tier 1 Storage with assured quality of service that comes from a shared storage system. SDS as processing power and server density has reached new levels. Offering lower costs, leveraging industry standard server hardware and the ability to colocate applications, storage servers, and hypervisors on the same CPU and disks. The deployment and consumption models discussion has turned to the use of converged and hyperconverged systems as “easy-to-deploy-and-manage” building blocks for IT. Use of these integrated systems is on the rise to accelerate time to value and free up data center resources. However, most workloads do not scale with perfect ratios of compute and storage over time, so there is still a need to grow your compute, networking, and storage resources flexibly to maintain efficiency. You will need both approaches as you approach your business transformation. The real key is to embrace these changing vectors and make decision points that do not incrementally bring more complexity into what is already a very complex data center environment. This converged and hybrid infrastructure requires instant mobility of data and application logic to assign the right resource to the right workload at the right time, enabling real-time data services and on-demand business agility. The HPE Composable Infrastructure architecture addresses business demands by combining HPE Converged Infrastructure, software-defined management, and hyperconverged systems. This approach reduces costs, eliminates silos, and frees available compute, storage, and networking resources.

Major market trends New data models will drive the need for new storage technologies. The confluence of mobile applications, cloud computing, and the IoT is causing massive increases in the volume of data that needs to be moved, processed, and stored by IT. Not only is the volume of data exploding, but the data itself, sources for data and usage models are undergoing significant transformation. Storage systems designed for traditional applications might not be the best fit for new data-intensive workloads such as intelligent edge computing and big data analytics. These new workloads are driving the need for new technologies, products, and business models.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Public cloud storage will continue to grow (but does not meet all needs). The public cloud has removed the headache of buying and managing hardware with a model that allows IT to easily spin up storage resources in the cloud almost instantaneously. However, data sovereignty requirements, security concerns, and public cloud bandwidth costs for data-intensive workloads mean that some workloads will continue to live in on-premises data centers for the foreseeable future. With this in mind, many IT organizations are using hybrid cloud strategies, which enable them to access data across on-premise and public cloud infrastructures. Lines will continue to blur between private and public cloud storage with the release of Microsoft Azure Stack in 2017, and the availability of new Amazon AWS products such as Snowball Edge and Snowmobile. In addition, I expect to see additional tactics from traditional enterprise storage vendors to combat the threat of public cloud such as increased availability of pay-as-you-go storage consumption models. Enterprise IT organizations will also look to more broadly deploy cloud-to-cloud backup capabilities to reduce risk and protect data. Server-based storage will be used for bigger projects. With the rise in SDS, IT organizations are combining commodity server hardware with SDS stacks for their private cloud infrastructure to lower costs and streamline operations. Although traditional external storage arrays still make up the majority of the market, server-based storage made up 24% of enterprise storage shipments in Q3 2016, according to IDC (International Data Corporation—a market research company), and has been steadily growing at the expense of traditional arrays over the last couple of years. Instead of being relegated to smaller projects, I expect to see SDS becoming an integral part of big data initiatives, workloads that require object-based storage and other data-intensive mission-critical projects this year. I also expect to see more SDS products hit the market designed for a wider range of use cases. Flash will continue to pick up steam. As the cost of flash media declines, the capacity of flash increases and more advanced data reduction technologies become available, hybrid and all-flash arrays are beginning to take hold beyond high-performance computing and big data analytics workloads. According to IDC, revenue from all-flash arrays reached $1.1 billion in Q3 2016—over 12% of the total enterprise market—at a growth rate of over 60% versus the same quarter the previous year. In 2017 and beyond, I expect that flash will become the “new normal” for some applications. The speed of flash is also likely to drive adoption of faster Fibre Channel and Ethernet storage networking protocols. Over the next 5 years, some estimates indicate that half of all data centers will deploy all-flash arrays. Other new storage technologies will begin to have a meaningful impact. There are a number of new technologies in various stages of development that will be interesting to watch in 2017.

||||||||||||||||||||

||||||||||||||||||||

First, NVMe (Non-Volatile Memory Express—interface specification for accessing non-volatile storage over PCI Express) will become available in more server models and hybrid/all-flash arrays as the technology becomes more affordable and the ecosystem expands. Second, expect to see the use of containers for enterprise storage becomes more prevalent. Third, expect more IT organizations evaluate and adopt products with predictive storage analytics capabilities (for example, Nimble Storage). Fourth and further out on the horizon, storage class memory will become better defined, with productization likely in 2018+.

The HPE Vision The HPE Vision represents a logical transition from the Transformation Areas Transform, Protect, Empower, and Enable (the what) to HPE Solutions (the how). As shown in figure 1-13, the strategy is designed to help customers harness the right mix of hybrid IT to accelerate service delivery, build next-generation data foundations to accelerate time to insight, and power a new generation of digital experiences to create business value.

Figure 1-13 The HPE Vision

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

HPE Solutions

Figure 1-14 Hybrid IT, Intelligent Edge, and services The HPE strategy represents a logical transition from the Transformation Areas Transform, Protect, Empower, and Enable (the what) to Transformation Solutions (the how). The strategy is designed to help customers harness the right mix of hybrid IT to accelerate service delivery, build next-generation data foundations to accelerate time to insight, and power a new generation of digital experiences to create business value. First, the world is going to be hybrid, and HPE makes hybrid IT simple. This is achieved through offerings in the traditional data center, software-defined infrastructure, systems software, private cloud, and through public cloud partnerships. Second, HPE powers the Intelligent Edge through offerings in campus and branch networking via Aruba products and the Industrial IoT with products like Edgeline Converged Systems. Third, services are going to be more critical than ever. HPE has the expertise to make it happen and can accelerate our customers’ journey through our 22,000 Technology Services professionals. HPE can build customer solutions from the ground up and deliver world-class support, advisory, and consumption models. The first step to transform to a hybrid solution is usually from a position of traditional infrastructure. HPE provides solutions in a four-phase approach to transforming to a hybrid solution: Phase 1: Transition from traditional infrastructure

Technet24

||||||||||||||||||||

||||||||||||||||||||

Phase 2: Modernize and automate Phase 3: Establish software-defined cloud model Phase 4: Broker hybrid IT at scale Each phase addresses a different step toward a hybrid solution. Phase 1: Infrastructure Consolidation and Virtualization Phase 2: Introduce rapid provisioning Phase 3: Establish private cloud Phase 4: Move to Enterprise private/hybrid cloud Customer benefits include: Shifts IT Spend away from operations and focuses on hybrid infrastructure innovations. Simplifies IT through rapid deployment of pre-packaged, workload-optimized converged systems. For more information, see this video https://www.youtube.com/watch?v=nzOKtdrsZvA. Establishes the foundation for hybrid IT computing; becomes the service broker, decides how and where to deliver services. Enhances disaster recovery by moving workloads and data to new data centers. Drives higher efficiencies and effectiveness, leverages existing investments, reduces energy costs, and enhances data center capacity.

HPE Pointnext services HPE Pointnext services make it easy to accelerate digital transformation. HPE Pointnext employs 25,000 professionals around the globe to help customers design, deploy, and optimize hybrid IT infrastructure. These professionals collaborate on around 11,000 projects per year, and the experience they have gained, coupled with an extensive library of infrastructure blueprints and design tools, helps to reduce the risk of customers undergoing transformation. Working with a large cross-section of businesses and a rich ecosystem of partners, Pointnext professionals can help with advisory, professional, and operational services to find, implement, and support the optimal solution for customers’ business needs. As shown in figure 1-15, HPE Pointnext services include: Platform Consulting Services—HPE provides assistance in developing IT strategies

||||||||||||||||||||

||||||||||||||||||||

to modernize and migrate to an on-premises automated, digital platform. Data Center Facilities Consulting Services—HPE helps in determining the best data center strategy to provide the right-sized requirements and providing the right mix of hybrid IT. Cloud Services—HPE provides an extensive catalog of cloud services to help customers define their strategy and roadmap for the journey to the cloud. Storage Services—HPE Pointnext storage consultants analyze and design storage systems to improve efficiency, reduce costs, and apply scalability to meet business needs, while keeping it all safe. Network Services—HPE helps businesses plan and manage the lifecycle of services. This could be in deploying complex, business-critical solutions and helping to mitigate costly installation and configuration errors. Onsite services provide an end-to-end program that includes project management, implementation, and multivendor integration.

Figure 1-15 HPE Pointnext services

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE OneView HPE OneView is an infrastructure automation engine to simplify operations, increasing the speed of IT delivery for new applications and services. Through software-defined intelligence, OneView brings a new level of automation to infrastructure management by taking a template-driven approach to provisioning, updating, and integrating compute, storage, and networking infrastructure. Designed with a modern, standard-based API and supported by a large and growing partner ecosystem, HPE OneView also makes it easy to integrate powerful infrastructure automation into existing IT tools and processes. As shown in figure 1-16, take command with HPE OneView to: Deploy infrastructure faster: Template-based automation enables IT generalists to rapidly and reliably provision resources in response to application owner requirements. Simplify operations: Agentless monitoring, online firmware updates, and a new Global Dashboard deliver streamlined lifecycle operations at scale. Increase productivity: The unified API allows developers and ISVs (independent software vendors) to harness the power of HPE OneView unify infrastructure automation with application and IT service delivery.

Figure 1-16 HPE OneView OneView innovations provide you the industry’s best infrastructure management experience, simplifying operations for HPE BladeSystem, HPE ProLiant servers, 3PAR

||||||||||||||||||||

||||||||||||||||||||

storage, HPE Networking, and HPE ConvergedSystems. It is an essential ingredient in the HPE Hyperconverged 380 virtual machine vending environment and power management for the industry’s first Composable Infrastructure, HPE Synergy. By deploying HPE OneView today, you place your IT operations firmly on the path toward a composable future. With HPE OneView, infrastructure can be configured, monitored, updated, and repurposed with a single line of code, allowing IT teams to more effectively meet changing application needs. It also includes templates, so domain experts are able to model important settings in software such as RAID configuration, BIOS settings, firmware baseline, network uplinks and downlinks, and SAN storage volumes and zoning. The intuitive dashboard provides an at-a-glance view of the status of your servers, storage pools, and enclosures. Color-coded icons clearly show which systems are functioning properly and which ones need attention. HPE OneView orchestrates firmware updates across server enclosures and interconnect modules while reducing application downtime. HPE 3PAR StoreServ Storage is fully integrated with HPE OneView server profiles, which saves time by automating a policy-driven rollout of enterprise class storage resources. OneView connects with HPE Cloud Service Automation 10 (CSA) to facilitate provisioning of cloud services that contain physical and virtualized components. Via HPE OneView integration, CSA allows you to provision multinode virtualization clusters using Server Profile Templates. Easy firmware management with scheduled firmware and OS driver updates and unlimited storage space for Service Packs for ProLiant (SPP) in external repository. Extends composable storage management by supporting HPE StoreVirtual Storage family. Advanced management of HPE ProLiant DL and BL Gen10 Servers, HPE Apollo 2000, 4000, 6000 Gen10 and Gen9 Systems, and HPE ProLiant ML350 Gen9 Server. Basic (monitoring) support for remaining (Gen8 and Gen9) HPE ProLiant ML Servers and HPE Superdome X Servers. The HPE OneView Global Dashboard allows you to manage up to 6,400 nodes in the data center today and more in the future. HPE OneView delivers proactive insights into the health of your entire infrastructure. Customers can simply download and run the OneView Global Dashboard alongside HPE OneView, allowing them to manage their HPE BladeSystem c-Class, HPE ProLiant DL, HPE Hyper Converged 380, HPE Synergy, and HPE Apollo systems as Technet24

||||||||||||||||||||

||||||||||||||||||||

one infrastructure. Note For more information about OneView, Synergy, and the Composable Infrastructure, use the hyperlinks below. http://www.hpe.com/info/oneview https://www.hpe.com/us/en/solutions/infrastructure.html https://news.hpe.com/fact-sheet-hpe-synergy-and-hpe-oneview-enhancements/

Align storage goals with business goals As shown in figure 1-17, aligning storage goals with the company’s business goals helps identify what is needed.

Figure 1-17 Data storage survey results Some of the main reasons for a hybrid cloud environment might include: Reduce costs (cited by 69% of IT decision makers) Manage data growth (cited by 65%) Deliver service and applications faster (61%)

||||||||||||||||||||

||||||||||||||||||||

Support business agility/market responsiveness (58%) Focus on the business goals rather than the specific hardware or configurations. In this way, you are not going to lock yourself into a specific storage type and you can be more flexible in implementing the strategy.

How are IT and storage business demands being measured? To measure whether IT and storage solutions are meeting business demands, companies have traditionally reviewed availability, performance, and cost. As shown in figure 118, measuring cost includes metrics such as OpEx, CapEx, total cost of ownership (TCO), and return on investment (ROI).

Figure 1-18 IT and storage business demand measurements However, business demands are rapidly growing and expanding to include agility, a metric that is determined by time to provision; time to react; and time to adapt to availability, performance, and cost.

Technet24

||||||||||||||||||||

||||||||||||||||||||

The demand for instant access anywhere and anytime along with the shift toward mobility and “bring your own device” are shaping IT and storage requirements. These new demands are being driven by evolving business needs and social changes enabled by new technologies.

Moving to Composable Infrastructure A Composable Infrastructure facilitates the move to a continuous services and application delivery model and enables applications to be updated as needed, rather than just once or twice a year. IT operates in a manner like how a cloud provider handles lines of business and the extended enterprise. This type of framework consistently meets SLAs and provides the predictable performance needed to support core workloads. As shown in figure 1-19, HPE offers several solutions that provide customers with a bridge to Composable Infrastructure. For example, HPE developed ConvergedSystem and hyperconverged solutions to help customers respond more quickly to their demands.

Figure 1-19 The journey to Composable Infrastructure With the release of Synergy, customers can move to a fully Composable Infrastructure. Synergy greatly reduces the operational complexity of managing infrastructure and enables customers to accelerate application deployment. It is designed for today and architected to anticipate the evolution of technologies in the future.

||||||||||||||||||||

||||||||||||||||||||

Why HPE? Figure 1-20 shows several arguments for choosing HPE.

Figure 1-20 Why HPE? The competition also has many achievements that they would like to share with your customers. Although our achievements are fantastic, to remain number one, we need to make sure that we really understand our customers. Never underestimate the competition they have got where they are by being good at what they do, we need to respect this. We also need to stay at the top of our game and make sure that we do not become complacent. But we are not alone, there is strength in numbers, this ultimately helps us to meet the challenges that our customers are facing. Providing a resolution that incorporates multiple facets allowing us to be a solutions provider and not just a hardware provider. HPE has always built strong alliances and partnerships, Intel, Microsoft, VMware, Cisco to name but a few and going forward new alliances and partnerships are being forged. When Stephen Gillett took over as CIO of Starbucks in 2008, the company was going through one of its most difficult phases. Starbucks’ stock price had lost half its value in less than two years, and its same-store sales were declining. The company needed to do something dramatic to rev up growth, and its CIO stepped up to the challenge. In early 2009, Stephen Gillett pitched the idea of setting up a new

Technet24

||||||||||||||||||||

||||||||||||||||||||

business unit focused on digital innovation and customer engagement to the company’s executive council. Starbucks’ dramatic success is a powerful lesson in collaboration. As CIO, Gillett collaborated closely with Starbucks’ CMO Annie Young-Scrivner to develop technical as well as go-to-market strategies. In Gillett’s words, “Collaboration means IT is integrated with customer offerings and loyalty programs.” Source: Capgemini Consulting and MIT Center for Digital Business Consider the killer whale, a large powerful fish, but they do their best work when they work as a team. The amazing sight of watching them herd herring into a ball gives credence to well-established collaboration and team work.

Gartner Magic Quadrant for solid-state arrays 2016 vs 2017 According to Gartner within the next 12 months, solid-state arrays (SSAs) will improve in performance by a factor of 10 and double in density and cost effectiveness, thereby changing the dynamics of the storage market. The Gartner magic quadrant for solid-state storage is intended to help IT leaders better understand SSA vendors positioning in the market. Compare the positioning difference between 2016 and 2017 at figure 1-21 to see the significant improvements HPE has made in the SSD storage market.

||||||||||||||||||||

||||||||||||||||||||

Figure 1-21 Gartner Magic Quadrants for solid-state arrays Gartner statements from the report: HPE offers a vast product portfolio that delivers attractive low-cost, low-capacity entry points, which can also scale up and out to nearly 4 PB of raw capacity Fast validation and efficient use of advanced, enterprise SSD technology allow the company to drive a very aggressive pricing strategy combined with compelling operating expenditure (OpEx) benefits HPE 3PAR StoreServ provides proven enterprise services and familiarity across the platform that appeals to its existing customers and channel partners HPE expanded its existing 3PAR StoreServ portfolio with its acquisition of Nimble Storage on 17 April 2017. The result is two product lines: the 3PAR StoreServ products, which feature common architecture and a shared software stack with the hybrid arrays; and the Nimble Predictive AF Series. Considering the disruptive nature of SSAs and their impact on the general-purpose external controller–based (ECB) disk storage market, Gartner has elected to report only on vendors who qualify as an SSA vendor. We do not consider SSD-only generalpurpose disk array configurations in this research. To meet our inclusion criteria, SSA vendors must have a dedicated model and name for an SSA, generally available by 4 March 2017, and the product cannot be configured with hard-disk drives (HDDs) at any time. These systems typically (but do not always) include an OS and data management software optimized for solid-state technology.

Gartner Magic Quadrant for general-purpose disk arrays Figure 1-22 shows the Gartner-published Magic Quadrant for general-purpose disk arrays.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 1-22 Gartner Magic Quadrant for general-purpose disk arrays At over $15 billion annually, this remains the largest sector within the storage market. HPE is positioned in the Leader’s Quadrant for its Completeness of Vision and Ability to Execute.

||||||||||||||||||||

||||||||||||||||||||

Learning check What benefit is provided by HPE OneView for a Composable Infrastructure?

Name two key data center trends that customers are addressing today.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: HPE Solutions Instructions: Answer questions 1 and 2. Your customer is asking you about the first step in transform to a hybrid solution. You explain there are four different phases in transforming to a hybrid solution. What is the first transformation phase? What does this phase address? Below are some typical customer issues; what questions could you ask to identify the concerns below: My business is moving fast but my infrastructure is moving too slowly. I have a limited space and IT staff. The data center is running out of space, power, and cooling from the sprawl. I need to refresh my applications and infrastructure to support cloud computing.

Watch the OneView video at the link below: https://www.youtube.com/watch?v=oFmAO_g835w

HPE Converged Infrastructure HPE introduced the concept of a Converged Infrastructure because businesses that were trying to compete in the application-driven economy found their traditional business infrastructure struggling. Traditional infrastructure is stable and allows organizations to slowly (over longer, planned periods of time) roll out applications that support the business, such as online transaction processing (OLTP) applications and databases. Converged Infrastructure has been a blueprint for the data center of the future. It accelerates the provisioning of IT services and applications through shared pools of interoperable resources that are managed as a service. With storage interoperability across a wide range of storage vendors, you can discover heterogeneous physical and virtual storage investments. Converged Infrastructure provides better alignment with

||||||||||||||||||||

||||||||||||||||||||

growing business demands. As shown in figure 1-23, Converged Infrastructure helps customers to focus on: Innovation—Move IT expenditures away from operations and double the amount of resources focused on innovation to drive competitive advantage. Responsiveness—Simplify IT through automation, the cloud, and workload-optimized systems. Cloud—Establish the ideal foundation for cloud by determining how and where to deliver IT services. Security—Implement security solutions that are quicker to install and update, including filtering and blocking more than twice as many attacks. Disaster recovery—Enhance disaster recovery, simplify multisite failover, and reduce downtime to seconds. ROI—Double utilization, leverage existing investments, reduce energy costs, and triple data center capacity.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 1-23 Converged Infrastructure features

Key initiatives for long-term IT strategy Figure 1-24 shows key initiatives for long-term IT strategy.

||||||||||||||||||||

||||||||||||||||||||

Figure 1-24 IT initiatives Enterprises tend to buy a solution when they need it rather than working on initiatives (which, in the longer term, would show the benefit). However, it is the work of presales and sales to consult and help them to understand the benefits of working on initiatives and having a longer-term overall plan. Doing this will help them when they need to make a decision and prevent them from making “point-in-time” decisions that could lead to disparate silos of standalone decisions (which eventually lead to extremely inefficient IT). Consolidation decreases the complexity and amount of hardware. This reduces CapEx and OpEx. Virtualization is happening at the server and operating system levels, and in the areas of storage, the network, and on the desktop, driving up the efficiency of an infrastructure (again, CapEx and OpEx). HPE Converged Infrastructure accelerates the provisioning of IT services and applications through shared pools of interoperable resources. Today, only HPE enables businesses to start small and grow to enable new businesses at a pace to meet business demands.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE Converged Storage principles HPE Converged Storage is based on principles shown in figure 1-25: Storage virtualization—Deliver simplified management, higher utilization, efficiency, and agility by separating the logical and physical resources. Scale out storage—Provide scalable capacity, performance, and availability. Thin technologies/thin provisioning—Reduce TCO, OpEx, and CapEx by using virtualization technology to make storage resources more efficient. Multitenancy—Meet SLAs and deliver virtual private arrays for secure workload segregation. Autonomic management—Implement self-management and optimization without the need for administrator intervention. Storage tiering—Use combinations of I/O accelerators in the server through SSDs, hard drives, StoreOnce technologies, libraries, tapes, and archiving systems to place data to meet price, performance, and latency requirements. Deduplication/compression—Reduce data redundancy to minimize storage capacity requirements. Green storage—Implement technology to reduce power, cooling, and environmental impacts. Data and business protection—Eliminate data loss and provide continuous availability of data, exceeding recovery time objective (RTO) and recovery point objective (RPO) demands in SLAs. Federated storage—Move data between storage systems in a peer-to-peer relationship simply, dynamically, and nondisruptively to improve storage utilization, balance workloads, and ease storage migration. Note Federated storage is a collection of storage resources that uses distributed volume management to shift workloads from busy arrays to those with available capacity.

||||||||||||||||||||

||||||||||||||||||||

Figure 1-25 HPE Converged Storage principles

Converged Storage is designed to deliver ROI Converged Storage is a single approach to storage that spans from entry-level to highend systems. Designed for today’s unpredictable workloads and unconstrained growth, Converged Storage offers common data services for block, object, and file workloads and is optimized for primary storage as well as information retention, analytics, and protection. HPE Converged Storage solutions address customer concerns by delivering nextgeneration technologies that provide converged management, scaled-out and federated software, and standard x86-based platforms. Converged Storage provides: Return on information to get more value for better decisions—Scale capacity and performance with intelligent search and policy-based tiering. Integration and support for autonomy and third-party analytics engines deliver Big Data to customers. Return on infrastructure to lower costs and boost agility—Improve utilization through massive consolidation, federated data mobility, and polymorphic designs that enable multipurpose platforms. Return on individuals to save time and reduce complexity—Simplify with common services and management from entry to enterprise across primary storage, information

Technet24

||||||||||||||||||||

||||||||||||||||||||

retention, analytics, and protection. Converged Storage from HPE is designed to eliminate boundaries. Management is simplified because business needs are incorporated in the design, allowing the storage to work autonomically. Common storage management across devices is integrated into converged solutions and third-party tools, providing proactive guidance and resolution, a template-driven policy engine, insight into analytics and planning, improved device accessibility, and consistent delivery across devices. Physical and virtual storage appliances feature shared foundational technologies, common management, external integration, and a single stack from entry to enterprise.

HPE Synergy—The first platform architected for composability Synergy, shown at the figure 1-26, combines hardware infrastructure, software, and services to deliver a single Composable Infrastructure platform that enables customers to be well positioned for the cloud. The Synergy platform reduces operational complexity for traditional workloads and increases operational speed for emerging applications and services. Using a single interface, Synergy composes physical and virtual compute, storage, and fabric pools into any configuration for any application. The extensible platform easily enables a broad range of applications and operational models such as virtualization, hybrid cloud, and DevOps.

Figure 1-26 HPE Synergy Converged Infrastructure eliminates server, storage, and networking silos and manages

||||||||||||||||||||

||||||||||||||||||||

resources with a single management tool. Composable Infrastructure is a single, fluid infrastructure that is optimized to deliver infrastructure for both traditional and new applications in seconds. It eliminates the high costs of overprovisioning and stranding of resources with built-in infrastructure intelligence, dramatically reducing CapEx. Synergy offers the following benefits of a Composable Infrastructure: Composable compute provides performance, scalability, density optimization, storage simplicity, and configuration flexibility. HPE Synergy Composer uses integrated, software-defined intelligence to self-discover, auto-integrate, provision, and scale from racks to rows. Its composable frame provides everything needed to run applications and allows IT to be quickly set up and consumed. Rack-scale, multifabric connectivity eliminates stand-alone top-of-rack (ToR) switches. High-density, integrated storage can be used to compose any compute with any storage pool (SDS, DAS, and SAN). Synergy is a single management interface that allows IT to instantly assemble and reassemble resources in any configuration. Through its single interface, Synergy composes physical and virtual compute, storage, and fabric pools into any configuration for any application.

Synergy composable storage Synergy composable storage creates a pool of flexible storage capacity using multiple storage options that can be configured almost instantly to rapidly provision infrastructure for a broad range of applications and workloads. Synergy storage is the industry’s first integrated storage solution capable of supporting Composable Infrastructure with compelling price-performance and high-density levels. It is composed along with compute and fabric resources to meet both current and future workloads. As shown in figure 1-27, Synergy composable storage offers high-density options that span from fully integrated direct-attached internal Synergy storage modules and SDS using HPE StoreVirtual VSA, to local drives and controllers installed in the compute modules, to fully orchestrated flash arrays with HPE 3PAR StoreServ.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 1-27 Synergy composable storage It enables you to respond to unpredictable demands regardless of the data type, connectivity protocol, or service-level requirement. As a key building block for many applications, Synergy composable storage is designed to store and share anything—file, block, and object data—with enterprise-class reliability and nondisruptive change management. Synergy eliminates hardware and operational complexity, so IT can deliver infrastructure to applications faster and with greater precision and flexibility. Synergy eliminates the high costs of overprovisioning and stranding of resources with built-in infrastructure intelligence, dramatically reducing CapEx. With improved economics, businesses can drive continuous service delivery during workload peak times and significantly reduce risk.

Learning check List three ways in which Converged Storage delivers a positive ROI.

Write a summary of the key concepts you learned in this chapter.

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Summary HPE Storage meets the needs of the Idea Economy. HPE solutions provide stepping stones from a traditional to a Composable Infrastructure. To keep up with storage trends, business is moving toward automated cost-optimized solutions. HPE Composable Infrastructure is designed around fluid resource pools, softwaredefined intelligence, and a unified API. HPE Synergy allows businesses to reduce CapEx with improved efficient economics.

||||||||||||||||||||

||||||||||||||||||||

2

HPE Storage Portfolio

Technet24

||||||||||||||||||||

||||||||||||||||||||

LEARNING OBJECTIVES After completing this chapter, you should be able to: Identify the Hewlett Packard Enterprise (HPE) Storage portfolio. Explore HPE Storage management tools. Summarize HPE Software. Discuss HPE 3PAR StoreServ operating system features.

||||||||||||||||||||

||||||||||||||||||||

Prelearning check Flash is all about consolidating racks and racks of storage. You need density to do that, and we are already pushing the boundaries there by being first to market with larger and larger solid-state drives (SSDs). Right now, you can do over 15 terabytes (TBs) on a single SSD, but if you want to continue to grow this density, eventually you would sacrifice performance for more capacity. There are 2 new flash technologies that will help us achieve this. Can you name them?

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE Storage portfolio The HPE Storage portfolio includes converged storage and traditional storage. HPE supports a range of Fibre Channel (FC)- and IP-attached (IP—Internet Protocol; storage attached over IP network) primary storage arrays including the latest addition, Nimble Storage family, as shown in Figure 2-1. Entry-level storage supports both block and file provisioning for network attached storage (NAS) and storage area network (SAN) environments.

Figure 2-1 HPE Storage portfolio Virtual Storage Appliance (VSA) compliments the portfolio with software-defined storage for converged and composable solutions. StoreOnce and StoreEver make up the portfolio providing the backup and archiving capabilities required in today’s SAN. In addition, state of the art storage networking products and accessories for entry-level, midrange, and high-end environments, including FC switches, converged protocol switches and directors as well as routers, host bus adapter (HBA), CNAs, transceivers,

||||||||||||||||||||

||||||||||||||||||||

and cables.

HPE Storage positioning As shown in Figure 2-2, HPE has flash solutions for our smallest to our most demanding customers. MSA—It is an affordable solution but with fewer data services than our other products. For smaller deployments or customers who need to consolidate just a few applications and for whom price performance is paramount. SimpliVity—From Hyperconverged to Composable 3PAR and Nimble—And the most advanced flash portfolio to provide a foundation for Hybrid IT. Integration with VMware, Microsoft, OpenStack, Docker, Mesosphere, Ansible, Chef, Puppet, Jenkins. Proven 99.9999% Availability. XP7—For customers who need extreme availability providing 14 nines of availability and includes mainframe connectivity.

Figure 2-2 HPE Storage portfolio More information related to comparison of HPE Primary storage products is shown in Figure 2-3.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-3 Comparison of HPE Storage products

HPE SimpliVity 380 Gen10 HPE SimpliVity 380, shown in Figure 2-4, available on HPE ProLiant DL380 Gen10 Servers, is a compact, scalable 2U rack-mounted building block that delivers server, storage, and storage networking services. Adaptable for diverse virtualized workloads, the secure 2U HPE ProLiant DL380 Gen10 delivers world-class performance with the right balance of expandability and scalability. It also provides a complete set of advanced functionalities that enables dramatic improvements to the efficiency, management, protection, and performance—at a fraction of the cost and complexity of today’s traditional infrastructure stack.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-4 HPE SimpliVity 380 Gen10

HPE SimpliVity backups HPE SimpliVity VM-centric backup creates a full independent logical copy of the virtual machine. There is no link or dependency to the original VM; therefore, operations on the original VM do not affect the backup or the restored copy of the VM. HPE SimpliVity VM-centric backups are extremely quick because they were designed into the platform to keep backups simple while maintaining performance. The backup set is deduplicated and compressed within the HPE SimpliVity Data Virtualization Platform (DVP). With local backups, no data is moved; only a copy of the metadata is created with pointers to the original data. For remote backups, only unique blocks of data, which are not available at the remote location (not specific to the last VM backup, but unique blocks compared to all other data available at the remote data center) are copied to the HPE SimpliVity hyperconverged nodes in the remote data center.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Modernizing the infrastructure HPE SimpliVity combines the entire infrastructure into a simple, flexible building block, so your customers can reduce the cost and complexity of their IT environment and deliver the technology their organization’s needs. Instructions: Watch the “SimpliVity” systems/simplivity.html.

video

at:

https://www.hpe.com/uk/en/integrated-

What were the issues with legacy architectures of the customer? What were the reasons for selecting SimpliVity? Name several benefits SimpliVity brought to customer compared to original architecture. Notes: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________

||||||||||||||||||||

HPE StoreFabric The HPE StoreFabric portfolio includes: HPE FC switches –

Gen6 FC (32 Gb FC)

–

Protocol support—4/8/16/32 Gb/s FC

–

Ports—24 to 64; active ports vary by model.

Brocade FC switches –

Gen5 FC 16 Gb FC, embedded FC switches for HPE Synergy, or BladeSystem (cclass)

–

12-16 downlinks

–

Internal ports and 12 uplinks, active ports vary by model.

HPE SAN Director –

Gen6 B-series and Gen5 C-series

–

Ports—48-511 ports

–

32/16 Gb models available

–

Supports “six nines” availability

–

Redundant hot-swappable components

||||||||||||||||||||

HPE StoreFabric Storage Networking Advantage—From small and midsized operations to data centers and cloud, HPE StoreFabric, shown in Figure 2-5, has dynamic end-toend solutions that solve even your most frustrating storage networking challenges.

Figure 2-5 HPE StoreFabric product positioning

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE StoreFabric offers Gen6 and Gen5 (32/16 Gb) FC Storage Networking solutions that ensure that your SAN does not suffer from outdated Infrastructure that was not designed to meet the needs of server virtualization or flash-based storage. Include storage networking in all your storage discussions, virtualization, Flash, and big data are requiring faster networks, and 4 Gb FC is in its EOSL (End of Service Life), and 8 Gb FC is starting to EOL (End of Life) too.

Gen6 Fibre Channel Accelerate data access, adapt to evolving requirements, and drive always-on business operations with Gen6 FC Solutions. Today’s mission-critical storage environments require greater consistency, predictability, and performance. Brocade Gen6 FC solutions meet those requirements by accelerating data access, adapting to evolving requirements, and driving always-on business operations. The HPE Gen6 SN8600B Director combines innovative hardware, software, and built-in instrumentation to ensure the industry’s highest level of operational stability while redefine application performance for hyperscale, missioncritical storage. Gen6 FC benefits are shown in Figure 2-6.

Figure 2-6 Gen6 FC benefits

StoreFabric Gen6 Director switch The HPE SAN Director portfolio contains the existing HPE StoreFabric SN8000B Switch 16 Gb Switch and the recently introduced HPE StoreFabric SN8600B 32 Gb FC Switch, shown in Figure 2-7. The HPE StoreFabric SAN Director consists of an 8-slot 14U chassis and a 4-slot 8U chassis providing up to 384 Gen6 32 Gb FC ports and up to

||||||||||||||||||||

||||||||||||||||||||

32 QSFP-based ICLs (inter-chassis link) which are equivalent to 128 FC ports.

Figure 2-7 StoreFabric SN8600B 8-slot Director switch StoreFabric B-series Gen6 eliminates performance bottlenecks and provides performance at 32 Gbps in the SAN. Enhanced diagnostics tools like Fabric Vision and Gen6 with IO Insight provide realtime monitoring and simplified management to optimize application performance and accelerate problem resolution, reducing operational costs and improving business agility. Protect customers’ investment with Gen7 FC-ready storage networking platform.

HPE StoreEasy Technet24

||||||||||||||||||||

||||||||||||||||||||

StoreEasy best fits when the customer is looking for heterogeneous file sharing, home directory consolidation, file sync and share, and Remote Office/Branch Office (ROBO) environments. With the fifth generation of StoreEasy Storage family, you get up to 211% higher capacity, 21% lower $/gigabyte (GB), 2X data transfer speed, 61% increase in supported users, and up to 25X faster RAID rebuilds. HPE StoreEasy portfolio is shown in Figure 2-8.

Figure 2-8 HPE StoreEasy portfolio StoreEasy 1000 Storage supports large numbers of users concurrently and integrates easily into new and existing environments with a straightforward, consistent management experience for IT generalists. Data stored on the StoreEasy 1000 is secured with virus scanning, snapshots, file and/or disk encryption, and backup. StoreEasy 3000 Gateway Storage is a new breed of efficient, secure, and highly available storage to simply address the file storage challenges of customers’ medium to large business and branch office SAN. Helping consolidate file and block data by adding file services to FC, integrated Small Computer System Interface (iSCSI), or SAS-based SANs (SAS—Serial Attached SCSI, SCSI—Small Computer System Interface). This provides higher return on investment (ROI) for your SAN and is easier to maintain than separate file and block storage pools.

HPE StoreEasy 3850 WSS2016 Gateway Single Node

||||||||||||||||||||

||||||||||||||||||||

The HPE StoreEasy 3850 Gateway Storage is an optimized, efficient, secure, and highly available file services gateway to address the file storage challenges of customers’ medium to large organizations and their SAN environments. HPE StoreEasy 3850 Gateway Storage, built on industry-leading HPE ProLiant DNA and Microsoft Windows Storage Server, integrates easily into new and existing SAN environments with a straightforward, consistent management experience for IT generalists or storage administrators. Using the HPE StoreEasy 3850 Gateway, organizations can leverage unused block capacity for file serving and maximize their SAN investment. The HPE StoreEasy 3000 Gateway increases uptime with standard features as near continuous health monitoring with HPE Active Health System, mirrored operating system SSDs, and advanced memory protection. Note The HPE Active Health System monitors and records changes in the server hardware and system configuration. The Active Health System assists in diagnosing problems and delivering rapid resolution when server failures occur. Confidently deploy large file systems with file system online self-healing that reduces the need to take systems offline, online verification, and precise and rapid repair that reduces downtime from file corruption from days or hours to minutes. StoreEasy protects data while at rest with file system encryption, and while it is being transferred with Server Message Block (SMB) encryption and signing. Cluster up to eight gateways for high availability and with transparent failover, enabling continuity of data access in the event of a failure or for online rolling maintenance updates. Protect data from simple user errors to catastrophic failure using StoreEasynative replication for data sets up to 100 TB. The HPE StoreEasy 3000 Gateway comes with preconfigured hardware and either Microsoft Windows Storage Server 2016 or 2012 R2. Get file gateway out of the box and online fast with tools that simplify installation and curb common configuration errors. HPE StoreEasy 3000 Gateway helps consolidate file and block data by adding file services to FC, iSCSI, or SAS-based SANs. This provides higher ROI for the SAN and is easier to maintain than separate file and block storage pools. The supported disk arrays are: HPE MSA Storage, HPE StoreVirtual Storage, HPE 3PAR StoreServ Storage, and HPE XP Storage.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Software-defined storage HPE StoreVirtual VSA, shown in Figure 2-9, sets up enterprise-level array functionality inside your server, delivering rich data services without dedicated storage hardware. Tight storage management integration with VMware®, Microsoft®, and Linux ensures easy deployment and data management on vSphere, Hyper-V, or KVM servers.

Figure 2-9 HPE StoreVirtual VSA software-defined storage HPE StoreOnce VSA is a virtual backup and deduplication appliance, excellent for backing up a smaller site or moving data from remote sites to a centralized data center. With HPE StoreOnce Catalyst, deduplicated data can be moved seamlessly across the enterprise without the need to rehydrate, and all data movement is controlled by your backup application from a single console. Tightly integrated with Veeam, HPE StoreVirtual VSA and HPE StoreOnce VSA provide built-in, enterprise-class data protection with granular recovery capability. HPE StoreVirtual VSA’s unique scale-out architecture lets you add storage capacity on the fly without compromising performance while its built-in high availability and disaster recovery features enhance business continuity for the entire server environment. HPE StoreOnce VSA—Hardware and hypervisor agnostic deduplication for backup delivered as a virtual machine. StoreOnce VSA provides virtualized environments with deduplication for backup, scalable capacity, set-and-forget backup scheduling, and centralized monitoring and management that requires no manual intervention.

||||||||||||||||||||

||||||||||||||||||||

Software-defined storage is storage functionality delivered as software. Instead of deploying dedicated storage hardware, you can consolidate shared storage onto the same servers as your applications. This easy-to-deploy storage solution converges applications and data on the same platform to reduce your hardware footprint by half without losing any required storage functionality.

HPE StoreOnce portfolio With the portfolio including not just dedicated hardware with the StoreOnce 3000 and 5000 series products, the StoreOnce VSA provides a cost-effective solution for virtualized environments that is fully compatible with the 3000, 5000, and 6000 series solutions enabling a fully automated backup solution to be deployed however broad your IT environment. With single pane of glass management across the entire backup infrastructure, the HPE StoreOnce solution provides a cost-effective, simple-to-deploy, configure, and manage backup solution. The HPE StoreOnce portfolio is shown in Figure 2-10.

Figure 2-10 HPE StoreOnce portfolio With the HPE StoreOnce 6600, you can back up huge amounts of data within short backup windows. High performance multistreaming and HPE StoreOnce Catalyst delivers backup speeds of up to 184* TB per hour ensuring that your data is always protected.

Technet24

||||||||||||||||||||

||||||||||||||||||||

StoreOnce single-node appliance Figure 2-11 shows StoreOnce single-node appliance features. Note Check the HPE StoreOnce Systems QuickSpecs for the latest numbers at https://h20195.www2.hpe.com/v2/GetDocument.aspx?docname=c04328820

Figure 2-11 StoreOnce single-node appliance features

StoreOnce single-node appliance—Key parameters Figure 2-12 shows StoreOnce single-node appliance key parameters. Note Fan-in and fan-out represent amounts of multiple sources or destinations mapped to appliance.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-12 StoreOnce single-node appliance key parameters

StoreOnce 6600—Key parameters Figure 2-13 shows StoreOnce 6600 key parameters.

Figure 2-13 StoreOnce 6600 key parameters

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE StoreEver With the broadest and most advanced portfolio in the industry, HPE StoreEver features support for LTO-8 (LTO—Liner Tape-Open, magnetic tape data storage technology). HPE StoreEver Tape provides tape storage that is critical to your customers’ comprehensive data protection and archiving. As the worldwide leader in tape drives, tape media, and tape automation, HPE StoreEver provides tape storage that is critical to comprehensive data protection and archiving. With a shelf life of 30 years, HPE StoreEver, shown in Figure 2-14, addresses all your long-term retention needs and delivers the most reliable long-term archival storage.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-14 HPE StoreEver portfolio HPE StoreEver LTO tape drives—High capacity and low cost of ownership, proven reliability, best practice media rotation, and cost-effective/durable archiving. HPE StoreEver Tape Autoloaders—The ideal solution for archiving and long-term data retention with industry-leading affordability, capacity, and security features (encryption and WORM—Write once read many).

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE StoreEver MSL Tape library—Unattended backup, disaster recovery, and archive solution with web-based remote management; the ideal solution for archive and longterm data retention due to industry-leading affordability; capacity, fast.

HPE T950 and TFinity ExaScale Edition Tape Library The HPE T950 tape library, shown in Figure 2-15, combines field-proven reliability with storage innovation, to give you access to a full suite of enterprise capabilities, including powerful library management, superior and proactive data integrity administration, and the best total cost of ownership of any tape solution.

Figure 2-15 HPE T950 and TFinity ExaScale Edition Tape Library HPE TFinity® ExaScale Edition Tape Libraries, shown in Figure 2-15, deliver enormous scalability to keep pace with unpredictable data growth. Choose from a range of base configurations and scale up to 53,460 LTO-8 tape cartridges, or 40,680 TS11x0 technology cartridges in increments of 10 LTO of 9 TS 11x0 Technology slots for capacity on demand. The HPE T950 can seamlessly scale up to 120 drives with over 10,000 media slots and can be configured either for LTO-6/7 and LTO-8 tape technology or TS11x0 Technology Drives. Mixed media support is not available. It will scale horizontally with linear

||||||||||||||||||||

||||||||||||||||||||

robots from 1 to 8 frames, with up to 120 drives to store a maximum of 190.3PB of enterprise data* and is designed to help organizations reduce risk and increase resilience for data integrity, data security, and high reliability. HPE’s TFinity ExaScale Edition Tape Library supports either LTO-6/7 and LTO-8 or TS11x0 technology or Oracle® T10000x drives and media, enabling customers to avoid vendor lock-in and select the tape technology that is perfect for their business needs. Incorporating between 1 and 144 tape drives, you can consolidate and store data using TS11x0 Technology or using LTO-8 tape cartridges. It is also equipped with dual robotics for high availability and bulk import and export Service Bay Frames for faster cartridge load and unload. Using innovative “shelves” instead of slots and TeraPack® containers in place of individual cartridges delivers up to a 50% reduction in data center floor space required versus competing offerings.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check An SMB customer has an entry-level SAN, and they ask you about archiving. What products would you discuss with them and why?

||||||||||||||||||||

||||||||||||||||||||

HPE primary storage Primary storage solutions from HPE, shown in Figure 2-16, are designed to enable a seamless transformation to modern architecture and protect existing investments. The solutions include: Tier 1 storage for mission-critical applications, virtualization, and cloud—High-end and midrange storage arrays are designed for IT-as-a-service. They leverage a single flash-optimized architecture with support for block, file, and object access. HPE offers fabric-attached primary storage to meet a variety of workloads. Entry storage for small and midsize businesses, along with remote offices and departments—Whether the customer is in start-up mode, evolving its infrastructure, or supporting departmental initiatives, entry storage from HPE can solve immediate storage challenges and lay the groundwork for accelerated IT growth.

Figure 2-16 HPE primary storage The HPE primary storage portfolio includes: HPE StoreVirtual HPE 3PAR StoreServ Latest to the HPE Storage family is the acquisition of Nimble Storage. Technet24

||||||||||||||||||||

||||||||||||||||||||

The Nimble Storage All-Flash Array is a scalable, cloud-ready platform. The arrays combine a flash-optimized architecture with HPE InfoSight Predictive Analytics to provide fast, reliable access to data.

HPE StoreVirtual HPE StoreVirtual solutions provide optimized storage for virtualized server environments. This technology allows thin-provisioned storage volumes and scale capacity on demand, helping to raise utilization levels and reduce energy consumption. As shown in Figure 2-17, the all-inclusive license of StoreVirtual covers enterpriseclass storage features, management, and value-add tools. Customers do not need to purchase additional software.

Figure 2-17 HPE StoreVirtual StoreVirtual SANs stripe and mirror multiple copies of data across a cluster of storage nodes, eliminating any single point of failure in the SAN. Applications have continuous data availability in the event of a power, network, disk, controller, or entire storage node failure. Double capacity utilization with reservation-less space-efficient snapshots and thin provisioning features are included at no extra cost. It provides high availability and 49% lower cost in small sites and remote offices.

HPE StoreVirtual 3200 The HPE StoreVirtual 3200, shown in Figure 2-18, is the ideal solution for the

||||||||||||||||||||

||||||||||||||||||||

following customers: Looking for an affordable, granular building block approach to scalable entry storage; has limited budget, cost is a key focal point. No on-site expertise, need for quick learning curve. Needing a storage solution that protects their investment and grows as the business grows. Plans for continued user growth at remote sites, looking to virtualize more applications at remote site, but has little or no experience at remote site, no dedicated IT staff handle all tasks. Looking for a unified solution with block and file.

Figure 2-18 HPE StoreVirtual 3200 StoreVirtual 3200 carries the legacy of the StoreVirtual brand forward. Designed for traditional spinning or newer hybrid configurations with the option to upgrade at any time to increase performance and capacity with best of both scale-up and scale-out features. With application-aware snapshots, thin provisioning, and optional Network RAID, along with a single, optional software suite, the StoreVirtual 3200 delivers advanced features such as Multi-site stretch cluster Adaptive Optimization and Remote Copy. More detailed specifications are shown in Figure 2-19.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-19 HPE StoreVirtual 3200 QuickSpecs StoreVirtual products provide scale-out storage platforms designed for the fluctuating needs of virtualized environments. These products are based on the LeftHand operating system. This solution delivers on availability, data mobility, disaster recovery, manageability, and growth. With inherent enterprise-class storage software functionality and leading virtualization software integration, StoreVirtual can support virtualization growth at all stages. Intuitive, common management simplifies storage administration. Additionally, storage federation enables data mobility across tiers and locations and between physical and virtual storage. StoreVirtual is also one of the most versatile storage platforms in the market today. Key features of StoreVirtual arrays include: iSCSI and FC connectivity—With iSCSI and optional FC connectivity, StoreVirtual works well even in mixed protocol environments. A single, flexible storage solution can be deployed for all locations. Customers can migrate from one protocol to the other without having to redesign or replace their storage. High data availability and superior disaster recovery—With nearly 100% high availability and reliability, the storage cluster can sustain multiple concurrent failures and still keep data online and accessible to applications. VMware-certified multisite disaster recovery delivers business continuity with failover that is transparent to users and applications. The multisite configuration maintains data availability beyond a single physical or logical site and validates full compatibility with VMware high availability

||||||||||||||||||||

||||||||||||||||||||

features. LeftHand Networks Remote Copy—Businesses can centralize backup and disaster recovery on a per-volume basis and use application-integrated snapshots for faster recovery. LeftHand Networks Remote Copy lets customers asynchronously replicate thin-provisioned, reservation-less snapshots between primary and remote locations. Peer Motion on HPE LeftHand Storage—With LeftHand Peer Motion, an online data mobility technology, customers can move volumes across all deployed StoreVirtual nodes regardless of platform or drive type. StoreVirtual is the only product that offers data mobility across tiers and locations and between virtual and physical storage. Adaptive Optimization for StoreVirtual—Adaptive Optimization is an auto-tiering technology for StoreVirtual that optimizes the cost and performance of StoreVirtual storage clusters. It balances data on volumes between tiers automatically. Beginning with LeftHand OS 11.0, this StoreServ feature is available for StoreVirtual systems. Adaptive Optimization enables you to install two types of IO (Input/Output) accelerators, or SSD, SAS, or SATA (Serial ATA; ATA— Advanced Technology Attachment, standard bus used for attaching drives to computer) drives, which makes two storage tiers available inside the StoreVirtual cluster. Scale and mix and match—Customers can scale from software-defined VSA to ProLiant rack and BladeSystem-based hardware models. They can mix and match product models to best fit their infrastructure. The result is simplified management and feature consistency. Space reclamation—Space reclamation allows LeftHand OS to make space available for new data storage by deleting files, moving data from one volume to another, and other activities that change the space used in the cluster. Note After space reclamation is enabled on a management group, that management group can no longer be downgraded to versions of the LeftHand OS that do not support space reclamation. Note that downgrading a management group is a supported function and not a typical user action for StoreVirtual storage.

Note For more information, check out the QuickSpecs https://www.hpe.com/h20195/V2/Getdocument.aspx?docname=c05158565

at

Technet24

||||||||||||||||||||

HPE StoreVirtual VSA HPE StoreVirtual VSA delivers all the benefits of traditional SAN storage without a physical SAN. StoreVirtual VSA allows you to build enterprise-level SAN functionality into a customer’s server infrastructure to deliver lower cost of ownership and easy management. Use VSA with SSDs to provide a high-performance storage solution. You can create an all-flash tier with VSA for maximum performance or use StoreVirtual Adaptive Optimization to automatically migrate only the data that needs the best performance to SSD. VSA runs on all ProLiant or third-party servers and BladeSystem server blades certified with VMware ESX or Microsoft Hyper-V. VSA allows businesses to consolidate storage virtual machines onto the same servers where application virtual machines reside, providing fully fault-tolerant server and storage environments. By totally removing a physical storage array, you can add hard disk drive (HDD) capacity to servers, along with VSA software, and reduce your upfront investment in new technologies by 80%. Converging server and storage technology into the same system cuts physical footprint for each solution in half. VSA provides the dedicated performance required for the customer’s primary site and delivers fully featured shared storage and disaster recovery for remote sites. With tens of thousands of systems deployed in production, VSA is the solution trusted by Fortune 500 companies and small businesses alike to deliver enterprise storage features without a SAN.

Hardware prerequisites HPE StoreVirtual VSA Software requires a vSphere-, Hyper-V-, or KVM-licensed server (KVM— Kernel-based Virtual Machine, Linux-based virtualization infrastructure) with the following: Server supported to run Hyper-V, vSphere, or KVM –

Microsoft Windows Server 2012, 2012 R2, 2016 (supported by LeftHand OS 12.7 or later)

–

VMware vSphere 5.x and higher, VMware vSphere 6.0 (supported by LeftHand OS 12.5 or later), and VMware vSphere 6.5 (supported by LeftHand OS 12.7 or later)

–

Linux CentOS 6.6/6.7/7.0, Red Hat Enterprise Linux 6.6/6.7/7.0, and UBUNTU 14.04

||||||||||||||||||||

Supported block storage option on Windows Server Catalog or VMware Compatibility Guide

||||||||||||||||||||

||||||||||||||||||||

Up to 50 TB total per StoreVirtual VSA (see license options below) Reservation of 2 virtual CPUs with 2GHz Memory requirements for StoreVirtual depend on virtualized storage capacity and use of Adaptive Optimization and/or Space Reclamation. A virtual switch with dedicated at least 1GbE connectivity. Full support for 10GbE. For detailed hardware prerequisites, go to http://www.hpe.com/storage/docs/storevirtual and review the “HPE StoreVirtual VSA Installation and Configuration Guide” document. HPE Storage recommends HPE ProLiant Servers for best results. What’s New

HPE StoreVirtual VSA Ready Nodes offer tested reference configurations with ProLiant DL servers. Integration with HPE Recovery Manager Central (RMC) for seamless data movement to StoreOnce Backup appliances & StoreOnce VSA, and bidirectional replication with 3PAR StoreServ arrays (Peer Copy). 10 TB and 50 TB no-term licenses with 3 years of 7x24 technical support embedded. Integration with HPE VM Explorer for centrally managed backup and replication for virtual environments.

Peer Copy from VSA to 3PAR As shown in Figure 2-20, Peer Copy allows creating bidirectional crash-consistent replicas between 3PAR and StoreVirtual VSA.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-20 Peer Copy Transport protocols—3PAR can be deployed with FC or iSCSI; VSA is deployed with iSCSI. Peer Copy is ideal for: Implementing DR and centralized backup for small remote sites Data migration between StoreVirtual and 3PAR Prerequisites include: RMC 4.0 base 3PAR OS 3.2.1 (with limitations), 3.2.2, and 3.3.1 StoreVirtual VSA OS 12.6

VSA Ready Nodes Realizing the benefits of software-defined storage and optimized outcomes with the flexibility of the HPE ProLiant server portfolio does not have to be challenging. With the new reference configurations for StoreVirtual VSA, virtualization solutions that converge compute and storage into a simple building block are made easy. StoreVirtual VSA Ready Nodes are HPE-tested configurations that provide customers and partners with a solid starting point for their project. The four StoreVirtual VSA Ready Node reference configurations target a range of use cases that can be addressed

||||||||||||||||||||

with StoreVirtual VSA: from small retail locations and larger remote offices to data centers with more demanding application and large-scale server virtualization. StoreVirtual VSA Ready Nodes support all StoreVirtual VSA features and deliver highly available storage within the same HPE ProLiant server that runs your virtualized workloads, converging the components of the solution stack. Whether it is a single-site cluster or stretched across racks or data centers, StoreVirtual VSA Ready Nodes simplify the process from selecting the server configuration to the roll-out of the virtualization solution. More details about the configurations can be found in the technical solution whitepaper via http://h20195.www2.hpe.com/v2/GetDocument.aspx?docname=a00000521ENW

HPE StoreVirtual DSM for Microsoft Multipath I/O( MPIO) HPE StoreVirtual DSM for Microsoft MPIO provides enhanced MPIO functionality: It offers automatic creation of an IO path to each storage system in the cluster on which the volume resides, plus a path for the administrative connection. It provides improved performance architecture over native MPIO solutions: –

Read IOs are always serviced by a storage system that holds a copy of the data being requested.

–

Write IOs are always serviced by a storage system that receives a copy of the data. Remaining copies (replicas) of the data are forwarded to the appropriate storage system based on the data protection level of the volume.

–

DSM for iSCSI: Can reduce latency and improve performance in Microsoft Windows environments. The DSM also optimizes data-path traffic in multisite (or stretch-cluster) configurations. Once the StoreVirtual DSM is installed on the Windows server, all supported StoreVirtual iSCSI volume connections made to an iSCSI SAN will be claimed by the StoreVirtual DSM. There is no need to configure the connections manually. When the StoreVirtual DSM is installed, all load balancing is automatically managed by path selection plug-in.

||||||||||||||||||||

Network RAID technology StoreVirtual SANs use Network RAID, which means that nodes can be split between locations to provide highly available storage that stays online during a site failure. Network RAID provides built-in synchronous data protection with automated failover and failback. It is configured on a per-volume basis and dictates how a logical volume’s Technet24

||||||||||||||||||||

||||||||||||||||||||

blocks are laid out, providing reliability and availability that can be configured to meet application and data requirements. Network RAID provides built-in, transparent, remote replication across storage systems, racks, rooms, buildings, and sites. Note StoreVirtual arrays with Network RAID bring low recovery point objective (RPO). However, the latency on the link needs to be considered when creating Network RAID across sites. Network RAID technology, shown in Figure 2-21, aggregates all resources and capacity into a single storage pool. So every volume in the cluster can access all available capacity. This also means that all cluster resources—CPUs, RAID controllers, network interface cards (NICs), caches, and disk spindles—can contribute to performance.

Figure 2-21 Network RAID technology To remove any single point of failure, the Network RAID functionality also stripes and protects multiple copies of data across a storage cluster. Redundancy is managed on a per-volume basis to improve storage utilization and match data protection of the volume to the application’s requirements.

Auto-tiering The HPE StoreVirtual 3200 Adaptive Optimization3 (AO) feature is an innovative technology that greatly increases the efficient use of faster storage devices (such as SSDs) by automatically and intelligently moving data between tiers of storage with different performance characteristics within a single storage system.

||||||||||||||||||||

||||||||||||||||||||

AO reduces planning complexity by using innovative algorithms to pinpoint hot data at a highly granular level and moving data in real time while maintaining a balance between background IO operations and application IO requirements. The result is a hands-free sub-LUN tiering solution that increases application performance substantially without any administrator intervention. AO is configurable on a per-volume basis and requires no tuning or specialized configuration to implement.

HPE StoreVirtual Peer Motion As shown in Figure 2-22, HPE StoreVirtual Peer Motion data mobility technology allows the movement of volumes across all deployed LeftHand/StoreVirtual nodes regardless of platform or drive type. StoreVirtual Peer Motion can also swap nodes in and out of a cluster, allowing for maintenance or upgrading to new technology without scheduled downtime or taking data offline.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-22 HPE StoreVirtual Peer Motion Peer Motion benefits include: Offers data and workload movement between storage clusters to address performance and capacity requirements Enables seamless LeftHand storage technology refreshes, eliminating downtime and/or service interruption during migration/upgrade activities

||||||||||||||||||||

||||||||||||||||||||

Promotes federated asset management by moving data from retiring storage arrays to new storage systems nondisruptively Allows flexible migration of data volumes between multisite clusters to align with the growing demand for cloud-based storage Enables clients to transparently move application workloads between storage tiers in virtualized and cloud computing environments Increases return on investment by combining HPE thin provisioning and volume migration technology, which lets users dynamically rebalance data volumes to the higher performance tier

HPE SmartClone volumes SmartClone is a StoreVirtual technology. SmartClone volumes are space-efficient copies of existing volumes or snapshots. They appear as multiple volumes that share a common snapshot, called a clone point. They share this snapshot data on the SAN. SmartClone volumes can be used to duplicate configurations or environments for widespread use, quickly and without consuming disk space for duplicated data. The SmartClone process can create up to 25 volumes in a single operation. Repeat the process to create more volumes, or use the Command Line Interface (CLI) to create larger quantities. SmartClone volumes can be created instantaneously and are fully featured, writable volumes. The only difference between regular volumes, snapshots, and SmartClone volumes is that SmartClone volumes depend on the clone point (the snapshot they are created from). Additionally, they might minimize space used on the SAN. For example, an administrator can create a volume with a specific operating system configuration. Then, using the SmartClone process, the administrator can create multiple volumes with access to that same operating system configuration, and yet only a single instance of the configuration is needed. Only as additional data is written to the different SmartClone volumes do those volumes consume additional space on the SAN. Multiple SmartClone volumes can be individually managed just like other volumes. SmartClone volumes can be used in long-term production environments. To create SmartClone volumes, there must be: A management group, cluster, and at least one volume Enough space on the SAN for the planned configuration The SmartClone feature is particularly beneficial for a virtual desktop infrastructure

Technet24

||||||||||||||||||||

||||||||||||||||||||

(VDI) implementation, because administrators can instantly produce many copies of a volume and allocate these copies to virtual desktops. All copies share any common data blocks, thus producing a highly efficient way of storing desktop data within the SAN. HPE Labs have observed that 3%–15% of a desktop system volume is typically consumed by unique data. SmartClones would be 90% efficient in a VDI environment, where the unique data written in each desktop is equal to 10%. A 250-desktop environment where each desktop was allocated 30 GB would normally require over 7.5 TB of usable storage. SmartClones could reduce the storage requirement in this example to less than 1 TB.

Application-consistent snapshots HPE StoreVirtual 3200 Storage allows for the creation of application-consistent snapshots of VMware® and Hyper-V VMs and of Microsoft® VSS-enabled applications. Application-consistent snapshots create instant application-consistent copies of data on a per-volume basis. Snapshots can be created in a variety of ways to meet business or application requirements. You can create snapshots manually, on a scheduled or scripted basis, or via the Microsoft VSS framework. You can then access snapshot data to recover individual files or folders from the volume or roll back an entire volume to a specific point in time.

||||||||||||||||||||

||||||||||||||||||||

Learning check Your customer wants to add more capacity to his StoreVirtual 3200. The StoreVirtual 3200 supports both small form factor (SFF) and large form factor (LFF) drive enclosures.

Technet24

||||||||||||||||||||

Activity: Using VSA Ready Nodes Instructions: Use the “HPE StoreVirtual VSA Ready Nodes: Reference configurations” documentation at: https://h20195.www2.hpe.com/V2/GetDocument.aspx? docname=a00000521ENW as necessary for this activity. Answer the questions that follow the scenario. –

Scenario: Your customer is looking at a VSA and wants a performance server for a medium hybrid virtualization project.

–

Which ProLiant server would you suggest?

–

How many and which disks are suggested?

–

What would be the expected IOPS for his 60/40 random read 8K workload on SSDs and HDDs?

||||||||||||||||||||

Notes: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________

||||||||||||||||||||

||||||||||||||||||||

________________________________________________________________________

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ HPE provides leadership in automation for 3PAR StoreServ customers by reducing the storage management burden by up to 90%. Major 3PAR StoreServ features are shown in Figure 2-23, Figure 2-24, and Figure 2-25.

Figure 2-23 HPE 3PAR StoreServ

Figure 2-24 3PAR data availability and protection

||||||||||||||||||||

||||||||||||||||||||

Figure 2-25 3PAR storage is efficient HPE provides leadership in data availability for 3PAR StoreServ customers by providing high availability, disaster recovery, and backup and restore solutions. HPE provides leadership in efficiency for 3PAR StoreServ customers by reducing TCO by >50%. Gartner Magic Quadrant states: HPE 3PAR StoreServ provides proven enterprise services and familiarity across the platform that appeals to its existing customers and channel partners.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ Family HPE 3PAR StoreServ Storage allows you to break down the silos that stand between you and the efficiency and agility required to succeed in the idea economy. HPE 3PAR StoreServ delivers rapid and automated provisioning, hardware-accelerated deduplication, and flash-optimized performance with sub-1ms latency within a single storage architecture. Combining all-flash capabilities and multiprotocol workload support with a converged management framework, HPE 3PAR StoreServ offers the industry’s most efficient storage engineered for the true convergence of block, file, and object access. Compared with the competition, HPE 3PAR StoreServ Storage systems are twice as easy to provision and twice as easy to monitor in one-third of the data center space. When customers apply space-efficient snapshot features such as those in StoreServ to the multiple full data copies used for developers and data warehouses, they can reduce capacity needs by six times, reducing cost per gigabyte. As shown in Figure 2-26, HPE 3PAR StoreServ product families include: HPE 3PAR StoreServ 8000—It delivers the performance advantages of a purpose-built, flash-optimized architecture without compromising resiliency, efficiency, or data mobility. HPE 3PAR StoreServ 8000 storage offers a range of options that support true convergence of block and file protocols, all-flash array performance, and the use of spinning media to further optimize costs. HPE 3PAR StoreServ 8000 storage offers rich, Tier 1 data services, quad-node resiliency, seamless data mobility between systems, high availability through a complete set of persistent technologies, and simple and efficient data protection with a flat backup to HPE StoreOnce backup appliances. The HPE 3PAR Gen5 Thin Express Application-Specific Integrated Circuit (ASIC) provides silicon-based hardware acceleration of thin technologies, including inline deduplication, to reduce acquisition and operational costs by up to 75% without compromising performance. HPE 3PAR StoreServ 9000—Is an enterprise-class flash array that helps you consolidate primary storage workloads—for file, block, and object—without compromising performance, scalability, data services, or resiliency. This newest 3PAR model based on the proven 3PAR architecture is purpose built for all-flash consolidation, delivering the performance, simplicity, and agility needed to support your hybrid IT environment. HPE 3PAR StoreServ 9450 Storage is available is an all-flash model, quad-node resiliency. Raw capacity of 6 tebibytes (TiB) with up to 576 SSF SSDs. Options include up to 80 16G FC ports or 40 10Gbe iSCSI/Fibre Channel over Ethernet (FCoE) ports or 12 10Gb/s Ethernet ports.

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ 20000—The HPE 3PAR StoreServ 20000 family offers enterprise flash arrays for demanding workloads with over 3.2 million IOPS at submillisecond latency and consumes 85% less space than traditional high-end arrays for massive footprint reduction. The StoreServ 20000 family scales out to eight nodes and is ideal for customers consolidating multiple racks of legacy, high-end storage. The StoreServ 20850 all-flash array can deliver over 75 GB/s of sustained throughput for maximum application performance. It also includes the StoreServ 20800 converged flash array, which scales up to 15 PB of usable capacity. This product family also features the HPE 3PAR Gen5 Thin Express ASIC.

Figure 2-26 HPE 3PAR StoreServ Family All models support unified block and file workloads as well as object access for clouddeveloped application consolidation. This density and workload flexibility assures that customers can prevent flash-array sprawl caused by introducing separate, capacitylimited flash architectures into the data center. Flash arrays and all-flash data centers must include built-in data availability and business continuity to reduce risk without performance penalty.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ 9450 and 20000 provide highest resiliency Figure 2-27 shows how hardware on HPE 3PAR StoreServ 9450 and 20000 is different from 8000: The 20000 has an average of about 26% faster rebuilds after a drive fails (based on 1.92 TB SSD). Power domains for compute and storage elements—isolates failure domains, which provides greater resiliency. Two ASICs and CPUs per node. Service all participating HBAs from front—service more easily, no built-in ports. Blue LED notification on front of enclosure that aid in serviceability on the 9450/20000 —blue: locate/safe to remove, green: ok, and amber: fault. Point-to-point SAS connection on the 20000 with directly connected drive enclosures will have a shorter pause in IO if backend path or enclosure fails as compared to a loop connection with daisy-chained enclosures.

Figure 2-27 HPE 3PAR StoreServ 9450 and 20000 resiliency

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ 9450 performance Figure 2-28 shows preliminary performance numbers coming from empirical measurements and estimates as of June 2017; final numbers are available in NinjaSTARS. Note For QuickSpecs on the HPE 3PAR 9450: https://www.hpe.com/h20195/v2/GetDocument.aspx?docname=a00005876enw

Figure 2-28 HPE 3PAR StoreServ 9450 performance

HPE 3PAR StoreServ 8000 HPE 3PAR StoreServ 8000 is the ideal solution for customers who: Need an affordably priced, nondisruptive solution that is scalable to 4 nodes. Looking for true convergence of file, block, and object. Require a single architecture to take advantage of common data services, including integrated file services, and common management across their data center. Deploy some or all applications and data types in physical or virtual environments and the cloud.

||||||||||||||||||||

||||||||||||||||||||

Need to upgrade their SAN, especially in virtualized environments, or to address unpredictable workloads with a modern architecture. Require low latency, higher speed support for higher VM density. Are large customers/enterprise customers needing departmental storage. Figure 2-29 compares different HPE 3PAR StoreServ 8000 products.

Figure 2-29 HPE 3PAR StoreServ 8000 product positioning Thanks to all-inclusive software licensing, 3PAR is the only array with included software that optimizes the entire IO path for flash at no additional cost: File Persona increases capacity requirements by opening up new file-based workloads for consolidation. Smart SAN simplifies StoreFabric attach with SAN zoning automation of 16/32 Gb switches. RMC drives StoreOnce attach with flash-integrated data protection. HPE 3PAR StoreServ delivers one architecture from midrange to enterprise, including an all-flash array with integrated file, block, and object access, as well as common data services and common management across the portfolio. Multitenant with support for multiple applications and workload in secure virtual domains Efficient with adaptive data reduction technologies (including deduplication and compression) that reduce capacity by over 75% Autonomic and self-configuring to improve administration efficiency

Technet24

||||||||||||||||||||

||||||||||||||||||||

Federated Peer-to-Peer transparent, nondisruptive data mobility Quality of service (QoS) to assure service level for most critical business applications

HPE 3PAR StoreServ—Converged flash 8440 and allflash 9450 HPE 3PAR StoreServ 8440—Converged flash array The performance of flash, the affordability of a hybrid storage with converged file, block, and object access and the scalability and resiliency of high-end storage. With up to four controllers and replication between any 3PAR StoreServ models, compromising between application availability and affordability is now a thing of the past.

HPE 3PAR StoreServ 9450 Storage—All-flash array Delivers close to twice the all-flash performance of the HPE 3PAR 8450 with 1.8 million IOPS at submillisecond latency, but it is also a consolidation up to18 PBs of usable storage. Add unified agility for any file or block workload, giving the customer up to 80 total host ports to maximize those unified capabilities. Additional differences are shown in Figure 2-30.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-30 HPE 3PAR StoreServ—Converged flash 8440 and all-flash 9450 Target customers include: Large-scale enterprise deployments where data center transformation is a priority. Customers wanting to move to an all-flash data center who want to consume storage in a secure on-premises model at cloud prices. Customers looking at consolidation of multiple workloads along with the best all-flash performance. Dell EMC, Pure, or IBM customers who are looking to modernize and consolidate their flash storage. Segments: Global, Commercial, and Enterprise accounts. Verticals: Financial, Healthcare, Manufacturing, and Service providers.

HPE 3PAR StoreServ 20000 Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ 20000 is the ideal solution for customers who: Need massive consolidation with workloads that require up to three million IOPS and/or scalability up to 9 PB raw. Have large mission-critical projects like ERP roll-out, private cloud. Next-gen data center requirements that require latest infrastructure, density, power, and cooling. Require complex disaster recovery and need to fan in or fan out multiple systems. Need to reduce the number of storage systems and consolidate/upgrade their legacy SANs. Additional details are shown in Figure 2-31.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-31 HPE 3PAR StoreServ 20000 HPE 3PAR StoreServ 20000 Storage is a new class of enterprise flash arrays for massive consolidation of your most demanding workloads. Hyper-scalable: Massive scalability with up to 8 Mesh-Active Clustered Nodes for up to 8 PB raw and 24 PB usable, to meet growing enterprise requirements. Flexible: Meet unpredictable business demands. Flash-optimized with greater than 3 million IOPS at less than 1 millisecond of latency, 40 GB/s bandwidth. Resilient: Consolidate with confidence, assure service levels with enterprise-class BC/DR using Async Streaming replication and assure end-to-end data integrity with

Technet24

||||||||||||||||||||

||||||||||||||||||||

Persistent Checksum.

HPE 3PAR StoreServ architecture Compared to traditional arrays where all advanced features have to be handled by the CPU in software, the 3PAR ASIC offloads the CPU in a way to allow much more advanced features to be used without scarifying system performance. Architecture of HPE 3PAR StoreServ 20000 is shown in Figure 2-32.

Figure 2-32 HPE 3PAR StoreServ 20000 architectural diagram HPE 3PAR StoreServ Storage is a family of storage systems with Tier 1 models that provide autonomic provisioning, tiering, and management capabilities. HPE offers a full range of services including multiple care levels and data migration, integration, and consulting services. This proactive approach leverages products, tools, and technologies to avoid problems and optimize performance. Multitenant and federated—These storage systems manage unpredictable workloads and virtual machine density. They support huge numbers of applications and mixed workloads without performance or resiliency impact. HPE Peer Motion software provides nondisruptive data mobility between federated systems. Efficient—Reduce upfront capacity requirements and technology refresh costs and eliminate stranded capacity using thin technologies. Autonomic—Customers can save up to 90% of administrator time. These systems simplify, automate, and expedite storage management intelligently and without administrator intervention. The system is self-managing, self-healing, and selfconfiguring.

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR ASIC HPE 3PAR StoreServ employs purpose-built HPE 3PAR ASICs at the center of each controller node. The ASICs feature efficient, silicon-based mechanisms to drive in-line deduplication. This implementation relies on the HPE 3PAR ASICs to generate and assign signatures to each unique incoming write request. HPE 3PAR 8000, 9000, and 20000 models are all built on the HPE 3PAR Gen5 Thin Express ASIC, which enables high levels of performance, scalability, and efficiency, including hardware-accelerated deduplication. The 3PAR Gen5 Thin Express ASIC provides silicon-based hardware acceleration of thin technologies, including in-line deduplication, to reduce acquisition and operational costs by up to 75% without compromising performance. Powered by the 3PAR Gen5 Thin Express ASIC, HPE 3PAR Persistent Checksum ensures end-to-end data integrity from the application server through the storage array for any workload. It remains completely transparent to those servers and applications. This data validation capability protects against flash media and transmission errors as data moves from ProLiant servers through the fabric and to the HPE 3PAR StoreServ Storage array.

Features of StoreServ architecture Key features of the StoreServ architecture include: HPE 3PAR provisioning technologies offer efficiency features for primary storage that can significantly reduce both capital and operational costs. Thin provisioning has achieved widespread adoption because it dramatically increases capacity efficiencies. Deduplication is also an essential consideration when looking into deploying workloads onto a flash tier or an all-flash array. Thin technologies can vary widely in how they are implemented. HPE 3PAR Adaptive Flash Cache (AFC) is a functionality of StoreServ that allows SSDs to act as Level 2 read cache holding random read data for spinning media that has aged out of dynamic random access memory (DRAM) read cache. AFC reduces application response time for read-intensive IO workloads and can improve write throughput in mixed-workload environment. AFC effectively increases the amount of random read data cached on high-speed media on a node. The HPE 3PAR File Persona Software Suite can be enabled on a StoreServ node pair with an optional license. It requires either a two-port 10GbE or a four-port 1GbE NIC installed in the system or the on-board 1GbE Remote Copy over IP (RCIP) port enabled

Technet24

||||||||||||||||||||

||||||||||||||||||||

for File Persona. The HPE 3PAR Data Optimization Software Suite v2 combines capabilities that provide autonomic storage tiering and dynamic data mobility with assured QoS. Adaptive Optimization provides highly reliable, nondisruptive, cost-optimized storage tiering at the subvolume level to deliver the right QoS to the right data at the right time on a large scale and Dynamic Optimization delivers it at the volume level. Peer Motion software enables seamless technology refresh and cost-optimized asset lifecycle management and lowers technology refresh capital expenditure. Priority Optimization software allows customers to ensure QoS on HPE 3PAR StoreServ Storage systems. With Priority Optimization, customers can protect their tenants and environments from unpredictable burst of IO and maintain predictable service-level agreement (SLA) for multiple tenants.

No single point of failure with StoreServ components StoreServ storage components such as storage nodes, cache cards, disk- and host-facing HBAs, power supplies, batteries, and disks feature N + 1 and in some cases N + 2 resilience. As a result, system interruption does not occur if any of these components fail. The system’s power domains are designed to prevent power loss to back-end disk devices by tolerating up to two disk chassis power supply failures. StoreServ storage systems offer up to four current load-balanced power distribution units (PDUs) per rack, providing a minimum of two and support for up to four separate data center power feeds to offer even more power redundancy and protection against events. All storage nodes run a separate copy of HPE 3PAR OS software. The software is both statefully managed and self-healing in the case of a failure or restart of a process, across all cache-coherent, active-active storage controller nodes. Each controller node supports one Ethernet connection to a switch or hub. Separate connections from the Ethernet switch or hub to at least two controller nodes are required to support redundancy. With redundancy, one IP address is shared between the two connections, and only one network connection is active at a time. If the active network connection fails, the IP address is automatically moved to the surviving network connection.

||||||||||||||||||||

||||||||||||||||||||

Notice At a minimum, the storage system requires one FC (or iSCSI) connection from a host computer to a controller node. HPE recommends separate connections from each host computer to each of the controller nodes in the storage system, with connections distributed evenly across all nodes.

Converged block, file, and object access 3PAR began life as a block device. And block is still the recommended mode for most applications including databases and virtualized environments. HPE included file capabilities starting with the 7000c models, and we call this File Persona. File Persona is integrated right onto the 3PAR controllers, and file and block are managed from a single pane of glass. All the current models can provide not only FC, iSCSI, and FCoE block access but also Network File System (NFS), Small Message Block (SMB), File Transfer Protocol (FTP), and REST API (REpresentational State Transfer Application Programming Interface) access. The initial use case for file was home directories and corporate shares. Those are still core use cases; however, HPE is continually qualifying additional use cases for File Persona. Block, file, and object access to HPE 3PAR StoreServ is shown in Figure 2-33.

Figure 2-33 Converged block, file, and object access

HPE 3PAR StoreServ: Embracing new technologies Technet24

||||||||||||||||||||

||||||||||||||||||||

3D NAND technology will enable SSDs to get cheaper, denser, and faster. The reason is that memory chips are stacked in layers to pack in more data, unlike single-plane chips currently used. This is cutting-edge technology, and 3PAR’s architecture enables customers to smoothly transition to 3D NAND. 3PAR arrays are built for IO not a particular drive type. That’s why 3PAR will continue to be on the leading edge of new drive types that bring down the cost of flash w/o passing on any of the risk thanks to our 5-year unconditional warranty. HPE 3PAR StoreServ benefits are shown in Figure 2-34. Note NAND flash uses floating-gate transistors. These are connected in a way that resembles a NAND gate: several transistors are connected in series, and the bit line is pulled low only if all the word lines are pulled high. In digital electronics, a NAND gate (negative-AND) is a logic gate which produces an output which is false only if all its inputs are true. (May 2018, Wikipedia.org)

Figure 2-34 HPE 3PAR StoreServ benefits

Adopting new flash Most architectures are designed for a single type of flash media and struggle to adopt new types of flash as they are released resulting in a lag before support.

The 3PAR architecture HPE 3PAR StoreServ’s unique architecture allows HPE 3PAR arrays to adopt new flash technologies as soon as they become available.

3D NAND For 3PAR, 3D NAND introduces new capabilities, including higher-performing writes, faster erase cycles, higher capacity, and improved endurance.

||||||||||||||||||||

||||||||||||||||||||

Architected for efficiency today and for what comes next Flash is all about consolidating racks and racks of storage. You need density to do that, and HPE has been pushing the boundaries by being first to market with larger and larger SSDs. Right now, you can do over 15 TB on a single SSD. But if you want to continue to grow this density, eventually you would sacrifice performance for more capacity. 3PAR 3D Cache, shown in Figure 2-35, with Storage Class Memory (SCM) and NonVolatile Memory Express (NVMe) solves this problem and ensures that we never have to make those tradeoffs.

Figure 2-35 HPE 3PAR efficiency

HPE 3PAR gets you ready for what is next There are two new technologies shown in Figure 2-36: Storage class memory or SCM and Non-Volatile Memory Express or NVMe. When you combine these with 3PAR’s intelligent caching, you get the next level of hyper-acceleration, called 3PAR 3D Cache.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-36 The next level of acceleration

Fifty percent lower average latencies and 80% higher IOPS Flash is all about consolidating racks and racks of storage. You need density to do that; we are already pushing the boundaries, there by being first to market with larger and larger SSDs. Right now, you can do over 15 TB on a single SSD, but if you want to continue to grow this density, eventually you would sacrifice performance for more capacity. 3PAR 3D Cache with SCM and NVMe can help to solve this problem. Any customers who purchase these 20000 systems along with the 9450 will be fully protected with NVMe and SCM support, as shown in Figure 2-37.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-37 3D Cache performance

HPE InfoSight for 3PAR HPE InfoSight is available to all customers, with an active support contract at no cost. Major features are shown in Figure 2-38.

Figure 2-38 HPE InfoSight for 3PAR InfoSight delivers prescriptive insights well beyond typical monitoring. One of the big limitations with monitoring is that it just creates more questions, whereas InfoSight applies data science and machine learning to give customers the answers they need, so they do not have to try and interpret graphs and log events anymore. For example, instead of just providing a typical latency curve, we will tell you the real

Technet24

||||||||||||||||||||

||||||||||||||||||||

impact of latency based on the underlining IO signature. So at the end of the day, it is a red or green light. HPE Storage and Nimble Storage users will login to the new HPE InfoSight portal using their “legacy” logins; over time (on password expiration), Nimble users will migrate to use HPE Passport login, new users will use HPE Passport. For HPE Storage users, StoreFront Remote (SFRM) pages will be embedded within the common HPE InfoSight portal along with the new cross stack analytics for 3PAR user experience. For Nimble Storage users, the Nimble InfoSight views will be available. For shared customers with both HPE Storage and Nimble Storage products, both perspectives will be available. Relevant help and support links will be provided to users in context of their HPE and/or Nimble perspective.

Portal navigation InfoSight provides analytics beyond the HPE 3PAR Storage system and allows HPE InfoSight to apply intelligent analysis throughout the full technology stack, from the VM down through the host, network, and storage array. Example of HPE InfoSight user interface is shown in Figure 2-39. Note Network and storage array performance attribution is not in the initial release. Data collection must be enabled via the 3PAR Service Processor.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-39 HPE InfoSight portal InfoSight collects and shows telemetry data only for the VMware ESXi servers connected to the HPE 3PAR and Nimble Storage system enabled by a 3PAR Service Processor. Multiple vCenter instances can be configured per 3PAR Service Processor/3PAR Storage system. Note VMware ESXi (formerly ESX) is an enterprise-class, type-1 hypervisor developed by VMware for deploying and serving virtual computers. As a type-1 hypervisor, ESXi is not a software application that is installed on an operating system (OS); instead, it includes and integrates vital OS components, such as a kernel. The name ESX originated as an abbreviation of Elastic Sky X. (May 2018, Wikipedia.org) The 3PAR Service Processor collects the following information: Datacenter to cluster to host mapping Host to storage path configuration Datastore to storage mapping VM to datastore mapping

Technet24

||||||||||||||||||||

||||||||||||||||||||

Multipathing and HBA settings and driver version information Performance counters for CPU, memory, disk access, and latencies Capacity information for disk objects

||||||||||||||||||||

Learning check Flash is all about consolidating racks and racks of storage. You need density to do that, and we are already pushing the boundaries, there by being first to market with larger and larger SSDs. Right now, you can do over 15 TB on a single SSD, but if you want to continue to grow this density, eventually you would sacrifice performance for more capacity. There are 2 new flash technologies that will help us achieve this. Can you name them?

Which requirements indicate you should propose an HPE 3PAR StoreServ 9450 rather than an 8450? (Select three.) A. Future growth requirements B. Limited budget C. High resiliency requirements D. High IOPS requirement E. Requirement for multiple data services such as File Persona and replication F. Gen5 ASIC capabilities required G. General purpose file requirements

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Activity: Understanding 3D Cache Instructions: Explore blogs listed below to find out more about 3D Cache. –

“NVMe and the need for ludicrous speed”: https://community.hpe.com/t5/Aroundthe-Storage-Block/NVMe-and-the-Need-for-Ludicrous-Speed/bap/6887059#.WRXBz-XyuUk

–

3PAR and Storage Class Memory: https://community.hpe.com/t5/Around-theStorage-Block/3PAR-and-Storage-Class-Memory-From-Ludicrous-Speed-toPlaid/ba-p/6900913#.WRXB5uXyuUk

||||||||||||||||||||

Watch the “HPE 3PAR 3D Cache for flash hyper acceleration” video at: https://www.youtube.com/watch?v=6RgMhfYZfso Notes: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

HPE Nimble Storage Nimble Storage All-Flash Arrays, portfolio shown in Figure 2-40, offer simplicity and flash performance within a scalable, cloud-ready platform. The arrays combine a flashoptimized architecture with HPE InfoSight Predictive Analytics to provide fast, reliable access to data and measured availability greater than 99.9999 percent.

Figure 2-40 HPE Nimble Storage

Technet24

||||||||||||||||||||

||||||||||||||||||||

Nimble All-Flash storage family Figure 2-41 shows the Nimble All-Flash storage family. The AF20Q and AF20 are the perfect entry points for all IT organizations that require speed and economy for performance-sensitive workloads. The AF40 and AF60 offer high performance and attractive economics for performancesensitive workloads that need the best blend of price, performance, and scalability. The AF80 is designed for consolidating multiple large-scale performance-sensitive applications with aggressive performance and high scalability demands.

Figure 2-41 Nimble All-Flash storage family

Scale out with clustered arrays Figure 2-42 shows clustering of up to four of any Nimble Storage arrays. Features include: Simple configuration Automated host connection management

||||||||||||||||||||

||||||||||||||||||||

Dynamic load-balancing and automated capacity rebalancing Linear performance scaling Automatic data migrations Add what you need, when you need it, avoid storage silos Maintain a single, unified storage platform Easier on storage budget Avoid overspending to achieve storage SLAs Extend storage ROI

Figure 2-42 Cluster of Nimble Storage arrays HPE Nimble O/S is the Operating System (firmware) that runs on all Nimble Arrays. The exact same versions of NimbleOS can run on all arrays.

Nimble Storage graphical user interface (GUI) Technet24

||||||||||||||||||||

||||||||||||||||||||

There are several methods for configuring and managing HPE Nimble Storage. They include: Array GUI: All functionality is available through the GUI (shown in Figure 2-43). CLI: All functionality is available through the CLI. API: All functionality is available through the API. There are also API plug-ins for vCenter.

Figure 2-43 Nimble Storage GUI Administrators can define user with different roles; once logged in from the home page, you can access performance metrics, capacity metrics, protection metrics, alarms and events, and also hardware information. From the interface, you can create volumes, clones, groups, and map volumes to hosts. Multipathing is supported for both FC and iSCSI.

||||||||||||||||||||

||||||||||||||||||||

HPE InfoSight HPE InfoSight provides customers with an automated and predictive capability for not only storage but across the infrastructure and into the virtualization (and application) layer. HPE InfoSight features are shown in Figure 2-44.

Figure 2-44 HPE InfoSight features

HPE InfoSight Predictive Analytics Proactive resolution: HPE InfoSight automatically predicts and resolves 86% of problems before you even know there is an issue. Solves storage and nonstorage problems: By collecting and correlating sensors across the infrastructure stack, HPE InfoSight uncovers problems spanning from storage to VMs. In fact, 54% of the problems HPE InfoSight resolves are outside of storage. Prevents known issues with infrastructure that learns: If a problem is detected in one system, HPE InfoSight begins to predict the issue and inoculate other systems. Every system gets smarter and more reliable through collective installed base insights. The support you have always wanted: Automation and proactive resolution put the focus on prevention, streamlining the process, and connecting you directly to support expertise. No more answering routine Technet24

||||||||||||||||||||

||||||||||||||||||||

support questions, sending log files, or attempting to recreate issues. HPE InfoSight answers questions like: What is your current growth rate? When will you likely need more storage capacity? How do you report back to the business? InfoSight accurately forecasts future capacity, performance, and bandwidth needs, especially for complex workloads like VDI.

Capacity trending and upgrade recommendations View of Trend Analysis shown in Figure 2-45—to help accurately forecast future capacity, performance, and bandwidth needs and to eliminate planning guesswork.

Figure 2-45 InfoSight capacity trending and upgrade recommendations

Benefits of radical simplicity are clear and measurable HPE InfoSight benefits are shown in Figure 2-46. As a result of HPE InfoSight, customers are able to overcome complexity and eliminate anxiety in three critical ways:

||||||||||||||||||||

||||||||||||||||||||

Figure 2-46 InfoSight benefits First, by predicting and preventing problems. The goal of predictive analytics is not just to identify customer issues and quickly resolve them but to use the power of the Nimble Storage globally connected installed base to prevent customers from having a problem to begin with. As HPE InfoSight analyzes the installed base, it is predicting and preventing in every customer environment. And if it uncovers an issue, it proactively resolves it. These are problems that often exist outside of storage. For instance, HPE prevented thousands of customers from experiencing issues across their networks, servers, and hypervisors. Second, HPE InfoSight sees what others cannot. So it provides clear insights up and down the infrastructure stack, across customer environments, into the past, and into the future. If you are trying to determine the root cause of a performance hit, you should not have to be a data scientist to effectively interpret the tools at your disposal. That is why HPE InfoSight has embedded data science and machine learning to give you the right answers using all the systems connected in the cloud. It also gives you access to infrastructure that gets smarter by learning from other systems. Third, because of HPE InfoSight, Nimble Storage is able to deliver a transformed support experience—one that customers love. In fact, HPE InfoSight automates the tasks handled by traditional Level 1 and Level 2 support staff. This has allowed Nimble to build a support organization made up entirely of Level 3 experts. So in the rare case that you need support, often in less than one minute, you will be connected directly with a Level 3 support engineer who will quickly resolve the problem. HPE support engineers are able to do this because they have all of your information at their fingertips. Technet24

||||||||||||||||||||

||||||||||||||||||||

Secondary Flash Arrays—Put your backup data to work Nimble Secondary Flash Arrays (SFAs) were built specifically to converge data protection and secondary storage workloads onto the same device. The Secondary Flash Array or “SFA” is the first secondary storage device to effectively use Flash. It provides the performance to enable instant restores and recovery, improving Service Levels, and Reducing Recovery Time Objectives (RTOs). As shown in Figure 2-47, SFAs give you the best of both worlds—by letting you radically simplify your infrastructure with data backup and disaster recovery, along with running valuable tasks such as Dev/Test and Analytics, all on the same storage platform.

Figure 2-47 Nimble Secondary Flash Arrays Built on Nimble Hybrid flash arrays, these secondary arrays offer always-on, inline data deduplication and compression, and HPE InfoSight Predictive Analytics. They deliver the performance, effective capacity, and availability and deep integration with Veeam to simplify data management.

Simplified data management The Nimble Storage SFA solution with Veeam simplifies data management through powerful capabilities that go above and beyond those of the array itself. The tight integration between the Nimble SFA and Veeam Backup & Recovery software gives backup administrators the ability to access Nimble Array capabilities by using the

||||||||||||||||||||

||||||||||||||||||||

Veeam UI. They can focus on using the software they know, without having to become experts on the storage hardware. The Nimble Storage SFA is engineered for high-efficiency backups, disaster recovery, and secondary data storage. Flash technology enables you to leverage backup data for development and testing, QA, and analytics and to instantly back up and recover data from any primary storage system. The integration of the SFA with the Veeam® backup software simplifies data lifecycle management and provides a path to cloud archiving. The combination of Veeam software and Nimble Storage flash hardware enables both near-instant backups and near-instant recovery directly from a storage snapshot, down to the granularity of a virtual machine (VM).

Twenty percent more usable capacity than other vendors Benefits of Nimble Arrays related to usable capacity are shown in Figure 2-48. How can Nimble deliver storage TCO that is 33%–66% less than other All-Flash arrays?

Figure 2-48 Nimble starts with 20% more usable capacity One of the reasons behind this is that Nimble Arrays start with 20% more usable capacity than all-flash arrays from other vendors. This comes from an architecture that was built for flash and therefore makes more efficient use of SSD capacity. This is even

Technet24

||||||||||||||||||||

||||||||||||||||||||

before we take into account further savings from deduplication and compression. For example, HPE compared Nimble to selected all-flash vendor: Out of 100 TiB raw, Nimble gets about 74 TiB usable. From the same amount of raw storage, other all-flash vendor gets only about 62 TiB. The astounding thing here is that we are able to achieve 20% more usable space, even though we have “Triple+” Parity RAID (many other vendors have only Double+ parity RAID) and we guarantee space for a spare. This means you get better TCO and better resiliency.

Viewing data reduction savings Notice at the top right corner of the screen you see the total data reduction savings as shown in Figure 2-49. By hovering your mouse over the Total Savings, you will see the data reduction savings for deduplication as well as clones and compression. The key callout here is that each data reduction method is listed separately so that your customer can see and measure the impact of each independently, allowing them to make adjustments as necessary to maximize performance and capacity. You can get further information by clicking on HPE InfoSight.

Figure 2-49 Data reduction reporting screen in a Nimble All-Flash Array.

||||||||||||||||||||

||||||||||||||||||||

As you can see in the lower right corner of the screen, in this particular case, 2 out of 2 volumes are participating in a File Server application category and their total savings are displayed. Again, the granular details about those savings can be obtained by hovering the cursor over the total savings number. It is important to note that savings are aggregated for all volumes with the same application category and reported together: Volumes with the same application category and block size are deduplicated together. Deduplication savings cannot be reported for individual volumes.

Triple parity protection plus spare drive Every chunk in a strip consists of data, metadata, and parity. This parity can be used to recover 1 sector corruption or IO error, when there is no redundancy in the stripe (3 DRIVES FAIL). Figure 2-50 shows the Triple+ Parity used by Nimble Arrays.

Figure 2-50 Triple parity protection The minimum AF-series array configuration consists of 24 drives. The distributed parity is represented by P,Q, and R in Figure 2-51. S is the distributed spare drive, as shown in Figure 2-51.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-51 Distributed parity

Radically simple “Timeless Storage” business model You get all-inclusive software licensing with Nimble, as shown in Figure 2-52. All software is included with the array. You have no decisions to make, nothing additional to buy.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-52 Nimble benefits HPE provides flat support pricing with an option for a free controller upgrade at the end of 3 years. Nimble Arrays can be upgraded nondisruptively without the need for forklift upgrades. HPE offers a satisfaction guarantee and futureproofing—the details of these are spelled out on the Nimble Storage website, including appropriate disclaimers around technologies such as NVMe and 3D XPoint.

HPE Cloud Volumes Notice HPE Cloud Volumes might not be available in all markets. An enterprise-grade multicloud storage service for running your applications in Amazon Web Services and Microsoft Azure. As shown in Figure 2-53, HPE Cloud Volumes are as easy to use as native cloud storage, but with enterprise-grade reliability and features your applications need. It is designed for easy data mobility, so you can have the freedom to move data between public clouds and your data center, and with cloudbased monitoring and predictive analytics you gain global visibility and insights across the stack, no matter where your data exists.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-53 HPE Cloud Volumes

Data protection Backing up and protecting data with HPE Cloud Volumes is easy, as shown in Figure 254. It has the ability to take instant snapshots (in seconds not hours), allowing you to protect your data quickly and more often. More importantly, given the efficiency of our snapshots, you do not need to make full copies and pay for them, but instead, you pay only for incrementally changed data that is typically just a few percent of your full data set. Cloud has enabled new uses for your data.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-54 HPE Cloud Volumes data protection HPE Cloud Volumes provide simple and fast copy data management that allows you to make instant copies of data for test/dev and analytics. This is much better than trying to make manual snapshots and copies which are slow and cumbersome with native cloud storage—and with HPE Cloud Volumes, you pay only for the newly changed data instead of full copies compared to native cloud storage.

Multicloud without lock in HPE Cloud Volumes also enable easy mobility between cloud providers, so you can use multiple clouds, as shown in Figure 2-55. You are no longer locked into one cloud, and when you decide to switch, there is no data migration or costly egress charges, instead you just flip the switch from the HPE Cloud Volumes portal, and automatically switches the connection to the new cloud provider instantly with moving a single byte of data!

Technet24

||||||||||||||||||||

Figure 2-55 HPE Cloud Volumes without lock in Plus, if you ever decide to move data off public cloud and back to your own data centers, you can do so easily without any egress charges.

Positioning Nimble within the HPE Storage portfolio Figure 2-56 shows positioning of Nimble Storage: Segments: SMB to Enterprise Workloads: VMWare vSphere, VDI, Collaboration applications such as Microsoft Office 365, All-flash performance needed for back office applications, SQL Server, and Oracle Buying triggers: Upgrading from traditional and legacy SAN storage to flash –

Needs greater performance for latency-sensitive apps, or faster performance for VM-intensive workloads.

–

Current infrastructure does not meet performance requirements.

–

Customer is concerned that on-premises purchases today might not fit with cloud strategy for tomorrow.

–

Looking for cost-effective all-flash storage modular building block for scale-out applications.

–

Needing a storage solution with powerful investment protection and the business grows.

||||||||||||||||||||

||||||||||||||||||||

–

||||||||||||||||||||

Wanting a highly available all-flash solution where they can start small and grow without disruption.

Figure 2-56 Positioning Nimble Storage

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Artificial intelligence for the data center Every second, HPE InfoSight analyzes and correlates millions of sensors from all of our globally deployed systems. HPE InfoSight continuously learns as it analyzes this data, making every system gets smarter and more reliable. Instructions: Watch the “Machine learning in https://www.hpe.com/us/en/storage/infosight.html.

the

cloud”

video

at:

How many problems can HPE InfoSight predict? What are the main benefits of HPE InfoSight? Notes: ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________

||||||||||||||||||||

||||||||||||||||||||

HPE management tools HPE management tools bring holistic approach to infrastructure management and allow radically simplified user experience for everyday tasks to accelerate IT processes. Various HPE management tools for storage products are shown in Figure 2-57.

Figure 2-57 HPE management tools Various management tasks for the 3PAR StoreServ array can be accomplished within each of the management tools above. The main tool to manage the array however is still the StoreServ Management Console (SSMC) GUI and the CLI. Aligned with HPE OneView, the dashboard leverages the same look and feel shared across HPE Converged System products, placing you one step closer to converged data center management.

HPE 3PAR StoreServ Storage management The default Dashboard design exhibits an easy flow and an informational overview of the HPE 3PAR StoreServ monitored structure. Each of the viewable donuts is direct-

Technet24

||||||||||||||||||||

||||||||||||||||||||

linked objects to monitored categories, as shown in Figure 2-58.

Figure 2-58 StoreServ management GUI A complex storage infrastructure can be frustrating and inefficient. The SSMC allows users to manage the entire storage environment from a single window. Through the dashboard view, you can assess what is happening across your data center within seconds. Whether you have two HPE 3PAR StoreServ Storage arrays or sixteen, the dashboard view is equally simple; it puts you just one click away from more detail.

Federation view in SSMC Federation view in SSMC, shown in Figure 2-59, enables administrators to manage multiple systems and: Keep thing simple. Balance Capacity and Performance. Helps provisioning new workloads. Federations make adding new systems easier.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-59 Federation view in SSMC Common provisioning groups (CPGs) are Policies that define Service and Availability level by: Drive type (SSD, FC—Fast Class, and NL—Nearline). Number of Drives (default = all; striping width). RAID level (RAID 10, RAID 50, and RAID 60) and set size. High availability level (HA magazine or cage). Multiple CPGs can be configured and optionally overlap the same drives. CPGs have many functions: They are the policies by which free Chunklets are assembled into logical disks (LDs). They are a container for existing volumes and used for reporting. They are the basis for service levels and our optimization products. As of 3PAR OS 3.3.1, RAID 60 is enforced for all FC and NL drives and recommended

Technet24

||||||||||||||||||||

||||||||||||||||||||

for SSD larger than 3.84 TB.

HPE 3PAR StoreServ CLI Many repeatable tasks can be completed through the CLI using the scheduler. The CLI is also able to complete some low-level tasks that are not available through the SSMC GUI. Example of the StoreServ CLI screen is shown in Figure 2-60.

Figure 2-60 StoreServ CLI For example, if you wanted to change the Step size associated with a CPG, you must do this through the CLI. Also when provisioning storage, the console presents you with a high-level view of your systems, including details on free capacity by storage tier. In cases where you are managing multiple systems, this view helps you choose the most appropriate system on which to configure storage for a new host or workload. VMware vSphere integration via plug-ins, hardware-assisted integration with Vmware vSphere APIs for Array Integration (VAAI), and support for VMware Virtual Volumes (VVols) as this new functionality becomes more broadly available from Vmware. Microsoft Windows integration lets you monitor the entire physical, logical, virtual storage, and cloud infrastructure via Microsoft System Center Operations Manager (SCOM) and System Center Virtual Machine Manager (SCVMM). Web Service (WS) APIs let you automate management of hosts, ports, volumes, CPGs, virtual logical unit numbers (VLUNs), storage space, and system information. OpenStack support over iSCSI and FC protocols provides the flexibility and cost effectiveness of a cloud-based open-source platform. HPE 3PAR CLI gives you powerful customization capabilities that are simple to configure and reduce the need for extra management tools and costly consulting

||||||||||||||||||||

||||||||||||||||||||

services.

HPE Smart SAN HPE Smart SAN, shown in Figure 2-61, takes you from: Manual provisioning using individual element managers for: Servers, SAN, Storage, many steps required and very time consuming

Figure 2-61 HPE Smart SAN To: Autonomic provisioning and SAN orchestration by 3PAR: Far less steps required, fast and easy service delivery within minutes, end-to-end in-band management, and online diagnostics Smart SAN introduces Target Driven Peer Zoning (TDPZ). TDPZ reduces the complexity of manually performing FC zoning tasks by automating these tasks through the HPE 3PAR Storage Array. The HPE 3PAR StoreServ Storage system communicates with the SAN fabric switches and creates the required TDPZs. TDPZ uses peer zoning to reduce the number of steps required to provision new storage by up to 90%. HPE Smart SAN is a new feature in 3PAR that enables SAN provisioning and orchestration by 3PAR array. Using HPE Smart SAN, customers will be able to very quickly, with little manual input, perform operations like: Zoning End-to-end QoS Security settings

Technet24

||||||||||||||||||||

||||||||||||||||||||

Diagnostics with self-healing capabilities

HPE Insight Remote Support (IRS) HPE IRS is a software solution that enables reactive and proactive remote support to improve the availability of supported servers, storage, networking, and printers. HPE IRS enables enhanced delivery of your warranty or support services contract. Benefits of HPE IRS are provided in Figure 2-62.

Figure 2-62 HPE Insight Remote Support IRS allows customers’ technology to “call home,” avoiding the need to log a case over the phone. This reduces the time it takes to resolve a call by up to 40% and enables a first-time fix in 95% of cases. IRS offers you a simpler way to manage your IT performance by automatically detecting problems and logging a support case with HPE. The tool delivers a real-time 24×7 view of your IT estate via the Insight Online Portal. IRS provides: Remote Monitoring—You can check the status of your IT environment via your smartphone or tablet device, anywhere, anytime. Automated Notifications—Your IT staff will spend less time performing physical device checks, troubleshooting, and logging support cases. Instead, IRS notifies HPE if a problem requires attention. Accurate Resolution—IRS has the ability to identify and resolve issues before they

||||||||||||||||||||

||||||||||||||||||||

cause unplanned downtime, resulting in your hardware being available for more time. HPE ProLiant Gen8 and Gen9, and Gen10, HP Integrity Superdome X servers. v7.x supports Solution products such as Converged Systems 300 and 700. Simplified deployment with 10-minute install and wizard-driven discovery and setup. Newer Hosting Device support including the latest Windows versions (such as Windows Server 2012) and broader virtualization support (such as VMware, Citrix, and Hyper-V). Support for newer browser versions of Microsoft Internet Explorer, Mozilla Firefox, and Google Chrome. Simplified device management; assign a contact name to up to 3,500 devices at one time. Personalized dashboard—a consolidated view of your devices using Insight Online on the HPE Support Center Portal, or the HPE Support Center Mobile App. Insight Online also provides the ability to manage service credits, partner access (when authorized by you), reports, contract and warranty management, support cases, and more. Active support, maintenance, and evolution of new features and capabilities in v7. Timely and secure Insight RS email updates with new SMTP authentication Standardized report experience Easier management of device groups and Custom Delivery identifications (CDIDs) Support for Windows Server 2016 for additional deployment flexibility

HPE Storage Operations Manager Whether your organization has a traditional data center or is part of a hybrid storage/cloud environment, it is important to have a comprehensive view of resources to better understand the implications of storage growth. Storage Operations Manager’s (SOM’s) dashboard views of storage capacity and key performance indicators (KPIs) benchmark performance data; identify trends, such as the time until you run out of storage for a volume or array; and respond faster to business needs. HPE SOM, shown in Figure 2-63, is a storage resource management (SRM) solution.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-63 HPE Storage Operations Manager With storage interoperability across a wide range of storage vendors, you can discover heterogeneous physical and virtual storage investments, centralize usage policies and reporting, view historical data trends, and forecast future requirements. SOM allows your storage investments to be in sync with overall performance objectives and business strategies. Flexible implementation options include inferred discovery, agentless, agent-rich, or a mixture of all to suit your business requirements. The storage environment ring is presented as a topology map that displays physical and virtual elements and the connections and path dependencies between devices. Its support for virtualization spans servers, fabric, and storage. As a result, you can achieve storage capacity and performance optimization across your storage infrastructure.

HPE Recovery Manager Central HPE RMC Software delivers unified data protection, addressing a range of RPO/RTO objectives with flexible recovery options using snapshots. As shown in Figure 2-64, RMC integrates 3PAR StoreServ primary storage and

||||||||||||||||||||

||||||||||||||||||||

StoreOnce Backup for converged data protection that delivers assured recovery of application-consistent recovery points with flexible recovery options. Features include: Optimized data movement: When a backup is triggered, RMC creates a snapshot on the 3PAR array. This is a simple process and decoupled from the actual movement of data, so the application only needs to be in backup mode for a brief instant. 3PAR RMC then moves the snapshot data—using the HPE StoreOnce Catalyst Protocol—to the StoreOnce Appliance. Unlike a traditional backup application, 3PAR RMC has full access to the 3PAR SnapDiff block-change detection technology. It therefore knows exactly what has changed since the last backup and only has to send those unique changed blocks at the array level across the network to the StoreOnce appliance. 3PAR RMC also ensures “low bandwidth transfer” by subdividing these blocks into 4KB chunks and checking with the StoreOnce appliance to see which blocks are new and need to be moved to the store. During this process, no host resources are used, virtually eliminating performance impact on the application. Full recovery at the cost of an incremental: 3PAR RMC differentiates from traditional backup solutions in the way the data is processed and stored by StoreOnce. When StoreOnce receives the changed blocks from 3PAR, the StoreOnce indexing process uses pointers to map previously written blocks along with the new ones so that a synthetic full copy of the data is always available for instant recovery whenever a restore is required. This is in direct contrast to certain traditional backup methods that expect you to recover a full backup and subsequently recover a series of incremental backups to restore the application data back to a certain point in time. Thus, 3PAR RMC provides the performance of incremental backups for what is essentially equivalent to a full backup as well as ensuring fast and simple recovery that saves the user a tremendous amount of time. Recover back from local and/or remote snapshots and/or backups to the original (parent volume or snapshot of parent volume) or a different 3PAR StoreServ array. Element Recovery Technology (ERT) allows more granular recovery of an individual VM or an individual file within a VM. Multistreaming support: Another key element of 3PAR RMC is the ability to split backups to StoreOnce into multiple streams to improve performance. By breaking up the application data into separate streams, 3PAR RMC can write concurrently to different objects within a StoreOnce Catalyst store, which helps combat the performance issues typically found with full backups. Multistreaming also helps with the delta-change “incrementals” in RMC synthetic fulls, in scenarios when high-change rates might mean significant numbers of changed blocks would be backed up.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-64 HPE Recovery Manager Central RMC for VMware (RMC-V) allows VM administrators to create, schedule, and manage VMware application-consistent snapshots on 3PAR StoreServ Storage arrays, StoreVirtual VSA, and the HC 250 Hyper-Converged system for VMware vSphere for rapid online recovery. Additionally, the RMC Express Protect feature enables automatic, fast, and efficient backup of VMware VMs and data stores directly from the supported primary storage platforms to StoreOnce Backup systems, independent of backup software. RMC for SQL (RMC-S) allows Microsoft SQL database administrators (DBAs) to create, schedule, and manage SQL application-consistent snapshots on a 3PAR array. RMC for Oracle (RMC-O) allows Oracle DBAs to create, schedule, and manage Oracle application-consistent snapshots on a 3PAR array. RMC for SAP HANA (RMC-SH) allows SAP HANA DBAs to create, schedule, and manage SAP HANA application-consistent snapshots on a 3PAR array.

||||||||||||||||||||

||||||||||||||||||||

Note SAP HANA is an in-memory, column-oriented, relational database management system developed and marketed by SAP SE company. SAP SE (stands for Systeme, Anwendungen und Produkte in der Datenverarbeitung) is a Germanbased European multinational software corporation that makes enterprise software to manage business operations and customer relations. (April 2018, Wikipedia.org) SE in the name of the company represents the legal form of the company: European Company (Societas Europaea). Protection of SAP HANA databases on HPE 3PAR StoreServ Storage through application-consistent snapshots on 3PAR and Express Protect Backups to HPE StoreOnce Backup systems. Protection of VMware VMs on HPE StoreVirtual VSA using application-consistent snapshots and Express Protect Backups to StoreOnce. Volume-level bidirectional data copy between HPE StoreVirtual VSA and HPE 3PAR StoreServ Storage array through HPE Peer Copy. Ability to run RMC Software on a Hyper-V VM. All RMC licenses are now included by default with all supported primary storage arrays including HPE 3PAR StoreServ 8000 and 20000 models, HPE StoreVirtual VSA, and HC 250 Hyper-Converged systems. Furthermore, the StoreOnce Catalyst Copy feature enables copying the database backups off-site to one or more StoreOnce appliances for DR purposes.

HPE OneView—Deploy infrastructure faster HPE OneView enables you to deploy infrastructure faster, saving valuable time and resources, using software-defined intelligence. IT specialists can create a catalog of workload-optimized infrastructure templates that allow IT generalists to rapidly and reliably provision and update compute, storage, and fabric resources in response to application owner requirements. HPE OneView enables faster compute automation through server profile templates that rapidly provision physical, virtual, and containerized systems including BIOS settings, local RAID configuration, firmware baseline, shared storage, and connections to the

Technet24

||||||||||||||||||||

||||||||||||||||||||

LAN and SAN. * When it is time to make changes, the change can be specified in the server profile template and then rolled out to multiple systems simultaneously, ensuring consistency and reliability. HPE OneView automation provisions storage volumes, zones the fiber channel SAN, and attaches 3PAR StoreServ Storage to the server profile. HPE OneView also proactively identifies SAN health issues from volume to target, which saves time and helps resolve SAN issues before they impact the business. HPE OneView connects BladeSystem infrastructure to the network in minutes instead of hours or days by automating the creation of Virtual Connect Ethernet and FC connections as well as uplinks to the LAN and SAN. Integration with HPE Networking’s Integrated Management Console (IMC) takes this process one step further. IMC listens for newly provisioned HPE BladeSystem enclosures and automatically connects them to the production network. HPE OneView is delivered as a virtual appliance, a preconfigured virtual machine ready to be deployed on a hypervisor host. User interface of HPE OneView is shown in Figure 2-65.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-65 HPE OneView HPE OneView is a scalable, resource-oriented solution focused on the entire life cycle —from initial configuration to on-going monitoring and maintenance—of both physical and logical resources: Physical resources are objects you can touch, such as server hardware, interconnects, top-of-rack switches, enclosures, storage systems, and racks. Logical resources are virtual objects such as templates or groups that, when applied to physical resources, provide a common structure across your data center. For example, server profile templates, logical interconnect groups, enclosure groups, server profiles, and volume templates are logical resources. Managed

If you add a managed server to HPE OneView, either in an enclosure or rack server, you can apply configurations, deploy server profiles, monitor operation status, collect statistics, and alert users to specific conditions. For more information, see "About managed c7000 enclosures" in the online help. Managing server hardware requires HPE OneView Advanced licensing. Monitored

If you add a monitored server to HPE OneView, either in an enclosure or rack server, you can monitor it for inventory and hardware status only. For more information, see “About monitored enclosures” in the online help. Monitoring server hardware uses a free license called HPE OneView Standard. Migrated

Enclosures from Virtual Connect Manager (VCM) and Virtual Connect Enterprise Manager (VCEM) can be migrated to HPE OneView with the configuration information so that the enclosure can be managed by HPE OneView. The managed enclosure requires HPE OneView Advanced licensing.

HPE OneView licensing HPE OneView supports the following license types: HPE OneView Standard—A free license that enables monitoring of supported HPE

Technet24

||||||||||||||||||||

hardware. HPE OneView Advanced—A license that enables all HPE OneView management features for supported HPE hardware. Includes an integrated Lights Out (iLO) Advanced license for the server hardware, which enables advanced management features. HPE OneView licensing is “per physical server” for managed nodes: There is no restriction on the number of VMs associated with a license. The HPE OneView virtual management appliance is available either by: –

Free download from HPE Software Depot at hpe.com/go/OneView/download, OR

–

Purchase of the HPE OneView Media Kit

||||||||||||||||||||

Best value is obtained by the customer when purchasing the HPE OneView software with HPE Gen10 servers or BladeSystem enclosures on the same purchase order. This also enables factory integration of the software with the hardware—factoryembedded software and licenses. Best Customer Experience can be obtained by purchasing standard factory-embedded licenses with servers/enclosures. Factory-embedded licenses provide true factory integration. For example, SKUs (stock keeping units) with “FIO” or “#0D1” in the description. Factory-embedded licenses are NOT available with G7 servers. Note For more information, consult the OneView deployment and management guide. https://h20565.www2.hpe.com/hpsc/doc/public/display?docId=c05270583

||||||||||||||||||||

||||||||||||||||||||

Learning check What two Express features are available with RMC?

HPE 3PAR StoreServ Operating System and Software HPE 3PAR All-inclusive Single-System Software is bundled with all new 8000, 9000, and 20000 HPE 3PAR arrays and is not available for purchase separately. The software titles included with the array are listed below. Single-System Software includes the following functionality and features: HPE 3PAR Operating System Software HPE 3PAR SSMC HPE 3PAR Virtual Copy, HPE 3PAR Dynamic Optimization, and HPE 3PAR Adaptive Optimization HPE 3PAR Priority Optimization, HPE 3PAR Virtual Domains, and HPE 3PAR Virtual Lock HPE 3PAR File Persona, HPE 3PAR Smart SAN, and HPE 3PAR Online Import HPE 3PAR Host Explorer and HPE 3PAR System Reporter HPE 3PAR Virtual Service Processor (not included in HPE 3PAR StoreServ 20000) HPE 3PAR RMC and HPE 3PAR Host Software HPE 3PAR RMC Application Suite

HPE 3PAR StoreServ Software portfolio HPE 3PAR provides all-inclusive licensing:

Technet24

||||||||||||||||||||

||||||||||||||||||||

It is bundled with all new 8000, 9000, and 20000 HPE 3PAR arrays. HPE 3PAR all-inclusive multisystem software offers enhanced capabilities, including Remote Copy capabilities and Storage Federation on 8000, 9000, and 20000 arrays. HPE 3PAR File Persona Software: Provides support for SMB protocol version 3.1.1. Native ACL (access control list) formats with native user/group IDs on disk. It provides file access auditing framework with integration into third-party ISV applications to meet enterprise security regulations.

HPE 3PAR 7/10000 StoreServ Software Suites HPE 3PAR 7000 and 10000 series arrays provide the opportunity to purchase software bundled into a series of suites designed to be affordable and simple to purchase. Overview of HPE 3PAR 7/10000 StoreServ Software Suites is provided in Figure 266. Select software titles can also be purchased separately, as indicated below.

Figure 2-66 HPE 3PAR 7/10000 StoreServ Software Suites

3PAR 8000 and 20000 software suites 3PAR 8000 and 20000 software suites dramatically simplify 3PAR software licensing, as shown in Figure 2-67. No more drive or capacity licensing and capping needed,

||||||||||||||||||||

licensing for all new 8000, 9000, and 20000 systems.

Figure 2-67 3PAR 8000 and 20000 software suites Required 3PAR OS releases: Any 3.2.2 (All single-system software except Online Import and RMC App Suite) 3.2.2 MU3 (All single-system software) 3.3.1 (All single-system software) Transition licenses for installed systems available

3PAR 8000, 9000, and 20000 software details Overview of 3PAR 8000, 9000, and 20000 software is provided in Figure 2-68. There are three categories of HPE 3PAR software: HPE 3PAR OS software—Core software runs on the system and delivers unique storage virtualization, virtual volume (VV) management, and RAID capabilities. HPE 3PAR host software—Host-based software products enable the system platform to address the needs of specific application environments, multipathing, and historical performance and capacity management. –

||||||||||||||||||||

HPE 3PAR Recovery Manager Software for VMware vSphere—This allows VMware administrators to create hundreds of virtual machine-aware snapshots and initiate rapid online recovery directly from within the VMware vCenter Server

Technet24

||||||||||||||||||||

virtualization management console. –

HPE 3PAR Recovery Manager for Microsoft Exchange—This extension to HPE 3PAR Virtual Copy intelligently creates and manages snapshots that can be used to quickly restore Exchange instances or databases (or nondisruptively back them up to tape) for near-continuous data protection.

–

HPE 3PAR Recovery Manager for Oracle—This extension to 3PAR Virtual Copy intelligently creates, manages, and presents time-consistent snapshot images of Oracle databases for rapid application recovery, near-continuous data protection, data sharing, and nondisruptive backup.

–

HPE 3PAR Recovery Manager for Microsoft SQL Server—This is another extension to 3PAR Virtual Copy that eases costs and administration by providing rapid, affordable online recovery of SQL Server databases from multiple, highly granular point-in-time snapshots. Businesses can quickly recover a database to a known point in time, speeding up a variety of operations including rapid recovery of the production SQL Server.

–

HPE StoreOnce RMC—RMC software protects business-critical applications without impacting performance. It integrates HPE 3PAR StoreServ primary storage with StoreOnce Backup systems to provide a converged availability and flat backup service that augments traditional backup processes. This automated software combines the simplicity and performance of snapshots with the reliability and costeffective retention of deduplicated backups.

–

HPE 3PAR Management Plug-In for VMware vCenter—This plug-in gives VMware administrators enhanced visibility into storage resources and precise insight into how individual virtual machines are mapped to data stores and individual storage volumes. When used with 3PAR Recovery Manager Software for VMware vSphere, it allows rapid online recovery from within the vCenter Server virtualization management console.

Additional software—Optional software offerings that run on the system provide enhanced capabilities: –

Virtual Domains—Provides access control to limit the privileges of administrators to subsets of volumes and hosts in a storage system. It ensures that VVs associated with a specific domain are not exported to hosts outside of that domain.

–

Virtual Lock—Enables you to specify the retention period for each volume or copy of a volume. Locking a VV prevents the volume from being deleted intentionally or unintentionally before the retention period elapses.

–

Virtual Copy—Enables instant virtual copy snapshots of existing volumes to be

||||||||||||||||||||

||||||||||||||||||||

created using copy-on-write technology so that virtual copies consume minimal capacity. –

Remote Copy—Enables data distribution and disaster recovery for applications through a host-independent, array-based data mirroring solution that can copy VVs from one storage system to a second one.

–

Peer Motion—Controls the migration of data from a source system to a destination system, enabling VVs and system configuration information to be copied to a new system without changing host configurations or interrupting access to the volumes.

–

Dynamic Optimization—Enables volumes to be dynamically tuned by changing volume parameters (for example, RAID levels, set sizes, and disk filters). Dynamic Optimization enables nondisruptive relayout of VVs. This functionality promotes system optimization through improved utilization of all physical resources present at a given time.

–

Adaptive Optimization—Takes a fine-grained, policy-driven approach to servicelevel optimization. It analyzes performance (access rates) for subvolume regions and then selects the regions with the highest IO rates. It uses the subvolume data movement engine built into HPE 3PAR OS to relocate those regions to the fastest storage tier available. It also moves less active regions to slower tiers to help ensure space availability for newly active regions.

||||||||||||||||||||

Important Customers who bought Remote Copy as an individual software title can buy a Remote Copy Extension Suite that includes only Peer Persistence and Cluster Extension (CLX) software.

Technet24

||||||||||||||||||||

Figure 2-68 3PAR 8000, 9000, and 20000 software details

All-inclusive licensing for existing 8k and 20k customers All-inclusive software licensing is available for the HPE 3PAR 8000 and 20000. There is no dependency on 3PAR OS release—it is supported with all versions of 3.2.2 and above. It is possible to add the all-inclusive license to an existing system with or without a hardware or software upgrade. Features include: Simple process via a “Transition License. Any 8k and 20k can be converted to a new model. Required 3PAR OS releases: –

Any 3.2.2 (All single-system software except Online Import and RMC App Suite).

–

3.2.2 MU3 (All single-system software).

–

3.3.1 (All single-system software).

||||||||||||||||||||

HPE 3PAR Adaptive Data Reduction With the advent of the latest 3PAR O/S release and all-flash arrays, we see support for some new data reduction technologies, as shown in Figure 2-69.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-69 HPE 3PAR Adaptive Data Reduction Thin-Provisioned volumes have been around for a while and can be re-thin using Zero Detect; the ASIC is responsible for detecting the incoming streams of zero data. This might require writing zero’s over deleted data or by using the Unmap feature of the O/S. The ASIC is also used to facilitate the inline deduplication that occurs for redundant data writes. HPE 3PAR deduplication relies on the ASIC to generate and assign a signature for each unique incoming write request. Express Indexing is then used to accelerate the signature comparison of new requests with existing data, with redundant data identified and prevented from being written to the SSDs. Instead of using dedicated engines within the ASIC to perform Compression, the HPE 3PAR ASIC is used to offload other resource-intensive operations from the CPUs, thus freeing them up to perform compression operations. The controller CPU is the main compression engine. Also, the system spreads the compression operations broadly across multiple CPU cores to expedite data compression. A unique HPE 3PAR technology called Express Scan improves compression efficiency by identifying incompressible streams of data and storing them in their native formats instead of wasting CPU cycles by attempting to compress data that is not compressible. It is important to understand that once data has been deduped and compressed, the result is a set of odd-sized pages in cache that are very inefficient to write to flash media. Data Packing takes these random-sized pages and packs them into small, fixed-size pages. This allows the system to attain a higher total system efficiency as compared to other all-flash platforms.

HPE Express Layout With the introduction of newer 3PAR O/S versions, Express Layout, shown in Figure 270, now removes the node-drive ownership paradigm seen on all monolithic, modular, and even modern all-flash arrays. In a traditional platform, a drive is owned by one controller with the secondary controller having ‘standby’ access to the drive. This Technet24

||||||||||||||||||||

||||||||||||||||||||

second controller cannot read or write to the drive, otherwise it will likely corrupt data on the drive.

Figure 2-70 HPE Express Layout In the HPE 3PAR StoreServ, each node in a node pair has active access to a drive, allowing both nodes in a node pair to read and write to the same drive at the same time. This allows for an architecture that is truly active/active from host port through to SSDs, enabling 3PAR systems to decrease the overheads associated with small RAID set sizes and drive individual SSDs to provide higher levels of performance.

HPE Express Writes HPE 3PAR Express Writes represents a series of HPE 3PAR OS protocol optimizations that optimize SCSI transactions, improving array CPU utilization and host-write latency. Introduced into 3PAR O/S 3.2.1, this new feature takes advantages of optimized writes to improve IOPS and reduce latency. The feature complements flash-optimized 3PAR arrays, but there is no exclusive requirement for SSDs. Express Writes is enabled with transparent FC protocol optimizations and ISCSI as of 3PAR O/S 3.3.1. Allows initiators to send Data along with the Command, rather than when the target requests it. Can significantly increase Write performance for small IOs. Eliminates 1xRTT per IO. Reduces interrupt and packet processing on both the Initiator and Target sides. These benefits extend to both spinning drives and flash-based media.

||||||||||||||||||||

||||||||||||||||||||

This is achieved through the HPE 3PAR OS by optimizing SCSI transactions and thus improving the number of interrupts per transaction, which results in improved array CPU utilization and lower host write latency.

HPE 3PAR Thin Technologies HPE 3PAR OS compaction technologies such as Thin Provisioning, Thin Deduplication, and Thin Reclamation offer efficiency benefits for primary storage that can significantly reduce both capital and operational costs. Thin technologies can vary widely in how they are implemented; some are complex to deploy, and others use coarse allocation units and cannot deliver the required space savings. HPE 3PAR Thin Technologies achieve advanced data compaction through leveraging built-in hardware capabilities and Express Indexing technology. HPE 3PAR Thin Technologies include: Thin Deduplication delivers in-line deduplication at scale via hardware acceleration, regardless of whether you choose to deploy an all-flash array or a flash storage tier on an existing 3PAR StoreServ Storage system. With this unique technology, businesses can increase usable capacity, lower total cost of ownership, and extend flash media life span with any 3PAR StoreServ Storage product. Thin Conversion, shown in Figure 2-71, provides preserved service levels at high performance during migrations. Thin Copy Reclamation keeps storage lean and efficient by reclaiming unused space resulting from the deletion of virtual copy snapshots associated with virtual copy and remote copy volumes. This delivers a tightly clustered, multicontroller, scale-out architecture. Thin Provisioning allocates VVs to application servers and provisions only a fraction of the physical storage behind these volumes. Thin Persistence eliminates capacity associated with data deletions in thin volumes. Inline zero detection eliminates free space with nominal impact on production volume performance. Users can run Thin Persistence on a regular basis without service-level concerns.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-71 HPE 3PAR Thin Conversion process

HPE 3PAR Thin Deduplication Data deduplication is a space reduction technique that identifies redundant/ duplicate data in physical storage and maintains only a single copy of data for all the redundant copies. HPE 3PAR Thin Deduplication software performs a full check on all data before marking it as duplicated, which is essential to ensuring data integrity for mission-

||||||||||||||||||||

||||||||||||||||||||

critical environments. To understand how this approach is unique and effective, it is important to understand how HPE 3PAR Thin Technologies drive data compaction through features that provide a broad range of capacity efficiency benefits. Thin Technologies leverage hardware acceleration to preserve high performance and ensure the resiliency demanded of Tier 1 storage. With Thin Deduplication software, the CPU-intensive jobs of calculating signatures for incoming data and verifying reads are offloaded to the ASICs, as shown in Figure 2-72, freeing up processor cycles to deliver advanced data services and service IO requests. This hardware-assisted approach enables in-line deduplication that increases capacity efficiency, flash performance protection, and flash media life span extension.

Figure 2-72 ASIC-based signature generation for in-line deduplication

HPE 3PAR Express Indexing Designed for mission-critical environments, HPE 3PAR StoreServ uses built-in, silicon-based mechanisms and Express Indexing, a mechanism that accelerates data signature comparison. Express Indexing is used for ultra-fast detection of duplicate write requests to preventing duplicate data from being written. Express Indexing protects flash performance and delivers extremely efficient, extremely granular blocklevel deduplication. When a new IO request is received, Express Indexing performs instant lookups using metadata tables to compare the signatures of the incoming request to signatures of data already stored in the array. When a match is found, Express Indexing flags the duplicate

Technet24

||||||||||||||||||||

||||||||||||||||||||

request and prevents it from being written to the back end. Instead, a pointer is added to the metadata table to reference the existing data blocks. To ensure data integrity, the software relies on the controller node ASICs to perform bit-to-bit comparison before any new write update is marked as a duplicate.

HPE 3PAR Data-at-Rest Encryption HPE 3PAR StoreServ arrays can optionally provide Data-at-Rest Encryption. This protects against data loss through theft, return of failed drives, or decommissioned arrays. This implementation uses a local key manager (LKM), and encryption is implemented at the physical disk level. The key must be backed up when it is created because it will be needed to unlock the drives in the event of a power cycle. Encryption is provided by use of self-encrypted drives. These drives can encrypt data at wirespeed. Data-at-Rest Encryption is a feature licensed at the frame level. The array must initially be set up for encryption. Self-encrypting drives (SED) and non-SEDs cannot be mixed in an array. An array is either encrypted or nonencrypted. Any drive that is removed or loses power for any reason must be unlocked to access the data on it. Some drives are FIPS 140-2 certified. As of HPE 3PAR OS 3.1.3, the LKM is not FIPS 140-2 certified, and there are no provisions to use an external key manager. From HPE 3PAR OS 3.2.1, use of external key managers and FIPS 140-2–certified key managers is possible. Note The Federal Information Processing Standard (FIPS) Publication 140-2, (FIPS PUB 140-2),[1][2] is a U.S. government computer security standard used to approve cryptographic modules. (April 2018, Wikipedia.org) As part of FIPS 140-2 compliance, starting from HPE 3PAR OS 3.2.1, HPE 3PAR supports an Enterprise Key Manager (EKM). EKM provides a complete security solution for unifying and automating an organization’s encryption controls by securely creating, protecting, serving, controlling, and auditing the encryption keys. HPE 3PAR OS 3.2.1 and newer supports Enterprise Secure Key Manager and SafeNet KeySecure k450 or k150. Both solutions support the HPE 3PAR StoreServ arrays that are supported for encryption. Both solutions that meet the NIST Key Management standards are validated for FIPS 140-2–level certification. Although there are other

||||||||||||||||||||

EKMs in the industry, HPE has only qualified these EKMs to function with the HPE 3PAR StoreServ array. Similar to the LKM, the EKM uses a single locking key for all drives in the array. The locking key is managed by the EKM and is not key manager sensitive, meaning whichever HPE key manager the user deploys, EKM will use the same methodology to provide a secure locking key. To protect the key, a new process “fipsvr” is deployed and is the only process that has access to the locking key. The key is only in memory during an operation, otherwise the key is stored in the EKM.

Workload-centric storage personas StoreServ storage is expressed by workload-centric storage personas. Storage personas comprise data access protocols and data services for the presentation of storage to hosts and clients. Specifically, HPE 3PAR StoreServ features two personas that are engineered into the core of the HPE 3PAR OS and system architecture and are managed seamlessly together through the HPE 3PAR SSMC and the scriptable HPE 3PAR CLI: HPE 3PAR Block Persona—HPE 3PAR StoreServ Storage is expressed by the Block Persona in the form of block volumes to server hosts via FC, iSCSI, and FCoE. Block Persona is ideal for: –

Virtualization

–

Database

–

Application workloads

HPE 3PAR File Persona—The HPE 3PAR File Persona can be enabled on an HPE 3PAR storage system node pair with an optional license. It requires either a two-port 10GbE or a four-port 1GbE NIC to be installed in the system or the on-board 1GbE RCIP port to be enabled for File Persona. File Persona extends the spectrum of primary storage workloads natively addressed by the system’s default Block Persona. Product highlights include: –

Rich file protocols including SMB 3.1.1, 3.0, 2.1, 2.0, and 1.0, NFSv4.0, v3.0, and FTP/FTPS to support a broad range of client operating systems.

–

NTFS security mode for near native Windows permission experience with crossprotocol locking complemented with static and dynamic user mapping.

||||||||||||||||||||

Technet24

||||||||||||||||||||

Note NTFS (“New Technology File System”) is a file system developed by Microsoft. –

Security of FIPS 140-2 validated Data-at-Rest Encryption as an optional additional measure to prevent unauthorized data access.

–

Transparent failover for clients via SMB 3.0 and NFS to allow for nondisruptive 3PAR OS upgrades or in the event of a controller failure.

–

Object Access API that enables programmatic data access via a REST API for select custom cloud applications from virtually any device anywhere.

||||||||||||||||||||

What is object storage? Unlike file or block storage systems, object storage does not use a conventional hierarchical structure. Data is instead split into objects that are stored across different servers providing built-in data protection for unrivalled data durability. Any data stored is assigned a Unique ID, and when a user or application puts the ID number in, their data is reassembled and retrieved from storage. This ticketing system allows data to be stored across multiple servers and retrieved quickly when needed— even if the servers are not in the same location. This helps improve collaboration across geographically dispersed data centers and offers built-in redundancy to improve data security and reliability. Most importantly, object storage can scale up beyond petabytes seamlessly. Through these storage personas, HPE 3PAR StoreServ provides truly converged block, file, and object access to simultaneously support an expanse of workloads and allows the best storage approach to be used for a given workload.

HPE 3PAR StoreServ File Persona The HPE 3PAR File Persona Software Suite is a licensed feature of HPE 3PAR OS that enables a rich set of file protocols and core file data services on an HPE 3PAR StoreServ system. File Persona extends the spectrum of primary storage workloads natively addressed by HPE 3PAR StoreServ from virtualization, databases, and applications based on the block persona. This includes client workloads such as home

||||||||||||||||||||

||||||||||||||||||||

directory consolidation, group and department shares, and corporate shares. The process involves converged controllers, agile capacity, and unified management. File Persona tightly integrates into the data center by supporting standard NAS protocols and file services such as authentication and authorization methods, antivirus servers, and variety of client operating systems. Most common terms used in association with HPE 3PAR StoreServ File Persona are shown in Figure 2-73.

Figure 2-73 HPE 3PAR StoreServ File Persona With the File Persona Software Suite, you can unlock the native file and object access capabilities within a StoreServ storage array, made possible by HPE 3PAR OS and the array’s converged controllers. By unlocking File Persona capabilities, you can provision file shares and block volumes from a single-user interface. This approach offers a unique solution that incorporates multiprotocol support into the system architecture to deliver a tightly integrated, truly converged solution for provisioning both block storage volumes and file shares from a single storage system. Unlike existing solutions, this truly converged solution extends the architectural benefits that the StoreServ storage system already delivers for block workloads to file shares and object access in a way that is simple to deploy and administer. Using File Persona within a StoreServ storage system enables uses such as: Enterprise file sync and share Home directory consolidation Technet24

||||||||||||||||||||

||||||||||||||||||||

Group/departmental shares Corporate shares Custom cloud applications

HPE 3PAR File Persona Features of 3PAR File Persona include: Maximum File Provisioning Group (FPG) size increased from 32 TiB to 64 TiB. Maximum number of files per FPG increased from 100 million to 250 million. Requires an extra 50 GB of dynamically provisioned swap space if offline filesystem check is required (350 GB total). Also added alerts and degraded state for FPG when exceeding this limit. Allows for higher port count on space-constrained 8000 series systems. No longer have to make a choice between iSCSI and file access for 2 node systems. Also enables replication over FC and peer persistence in combination with file access. As with the existing cards, mode 1 and mode 6 bonding are supported. 10GbE ports are also usable for remote copy, but both ports of the add in card must be assigned to either file persona or remote copy. Figure 2-74 shows how File Persona extends the spectrum of primary storage workloads natively addressed by 3PAR StoreServ from server workloads, such as virtualization, databases, and applications via the Block Persona to also include the client workloads of home directory consolidation, group and department shares, and corporate shares via the File Persona.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-74 HPE 3PAR File Persona features

HPE 3PAR File Persona logical architecture HPE 3PAR StoreServ File Persona Software Suite comprises the following managed objects: FPG—Is an instance of the HPE Adaptive File System. It controls how files are stored and retrieved. Each FPG is transparently constructed from one or multiple VVs and is the unit for replication and disaster recovery for File Persona. There are up to 16 FPGs supported on a node pair. FPG is the unit of failover; it holds virtual file server (VFS), share, and configuration information for easy data recovery; no auxiliary synchronization is required. There is no file system inconsistency because of unexpected power loss or node failures. VFS—Is conceptually like a server. It presents virtual IP addresses to clients, participates in user authentication services, and can have properties for such things as user/group quota management and antivirus policies. There are up to 16 VFSs supported on a node pair, one per FPG. File Stores—Are the slice of a VFS and FPG at which snapshots are taken, capacity quota management can be performed, and antivirus scan service policies customized. There are up to 256 File Stores supported on a node pair, 16 File Stores per VFS. File Shares—Are what provide data access to clients via SMB, NFS, and the Object Access API, subject to the share permissions applied to them. Multiple File Shares can

Technet24

||||||||||||||||||||

||||||||||||||||||||

be created for a File Store and at different directory levels within a File Store. In case of node/file persona failure, FPG, VFS, and all associated shares move to partner node in a failover pair. Failover is transparent for clients, depending on the protocol used (SMB 3.0, NFS). There is complete fault isolation from block services.

HPE 3PAR File Persona features Features of File Persona include: Disaster recovery across storage arrays—Disaster recovery uses standard HPE 3PAR Remote Copy to synchronize block-level data with other arrays. Remote Copy needs to be set up the same way as it would be with other 3PAR VVs. Failover between arrays—Failover between arrays requires manual administrator intervention and requires downtime. During the failover, network resources such as IP address, shares, and exports are not available and need to be restarted on the target array. Clients will need to “reconnect” to the file-sharing services after the failover is complete. Automatic, scheduled tasks such as antivirus scanning schedules, snapshot, and schedules need to be migrated manually. File-share backup process—A file-share backup process generally works with every independent software vendor. The backup server (disk agent) reads configuration and user data from the regular NFS or SMB shares. Before a backup is performed, the backupfsconf command is used to create the configuration data file. The .admin file store contains the configuration information data file, which needs to be protected as well. File-share restore process—The .admin file store share is created and recovered from backup. The configbackup file is used to recreate the other user data file stores and shares. User data is recovered to the recreated file shares.

HPE 3PAR Dynamic Optimization HPE 3PAR Dynamic Optimization, shown in Figure 2-75, is an autonomic storage tiering tool that provides the ability to react quickly and nondisruptively to changing application and infrastructure requirements. Dynamic Optimization enables organizations to achieve optimized data QoS levels at the lowest cost, at all times, across all stages of the disk-based data life cycle. It enables users to nondisruptively alter service levels associated with a storage volume (RAID level, subsystem failure protection level, drive type, stripe width, and radial placement) and to rebalance

||||||||||||||||||||

storage volumes, each with a single click. Users can initiate volume migration to the best tier, based on application needs or historical array performance analysis.

Figure 2-75 HPE 3PAR Dynamic Optimization and Adaptive Optimization Policy Advisor for Dynamic Optimization is an add-on application that analyzes how volumes on the HPE 3PAR storage system are using physical disk space and makes intelligent, autonomic adjustments to ensure optimal volume distribution and tiering of storage volumes. Dynamic Optimization: Aligns application requirements with data QoS levels flexibly, precisely, and on demand: –

Nondisruptively alters service levels of a volume with a single click, without scheduling, planning, or impacting application performance. Each storage volume resides on a single tier but maintains mobility between tiers throughout the application life cycle.

–

Drives large-scale performance optimizations with one-button system rebalance after adding new resources to the system or moving a large number of volumes between tiers.

||||||||||||||||||||

Technet24

||||||||||||||||||||

–

Can be used to seamlessly adjust the RAID level, subsystem failure protection level, drive type, stripe width, or radial placement of volumes to achieve optimized data QoS levels.

Allows users to react quickly to changing application and infrastructure requirements: –

Enables service-level change orders to be made on demand, aligning appropriate storage resources as needed with the changing value of data.

–

Autonomically and nondisruptively rebalances the entire system or multiple volumes with a single button based on configurable policies.

–

Eliminates the cost of overprovisioning to accommodate usage spikes as well as the time-consuming, manual migration of data to lower tiers to reduce storage costs.

||||||||||||||||||||

HPE 3PAR Adaptive Optimization Adaptive Optimization software, shown in Figure 2-75, is a granular, policy-driven, autonomic storage tiering solution that delivers service-level optimization for virtual and cloud data centers to reduce cost and increase agility and minimizing risk. By using Adaptive Optimization technology to migrate data between storage tiers within individual storage systems, StoreServ products provide a unique way to balance performance and capacity within a storage pool and lowering overall cost. Automated tiering intelligently tracks in-flight use statistics at a granular level and migrates highly accessed data to higher-performing storage devices in real time. An important balance between internal data movement and application IO is maintained using advanced comparative usage metrics to ensure that data is not moved unnecessarily or too often. Adaptive Optimization maintains a “heat map” for blocks on all Adaptive Optimization–permitted volumes and will monitor workloads continuously. Frequently accessed blocks are promoted to Tier 0 (faster tier), and less frequently accessed blocks are moved down to Tier 1 or Tier 2 (slower tier). This avoids thrashing and unnecessary data movement. The system ensures that Tier 0 is always close to 100% utilization. Disabling Adaptive Optimization on a volume causes blocks to be moved to a lower tier. Benefits of using Adaptive Optimization include: Service-level optimization delivers cost savings:

||||||||||||||||||||

–

Matches data at the subvolume level to the most cost-efficient resource capable of delivering the needed service level, so the right QoS is delivered to the right data at the right time.

–

Leverages SSDs to allow customers to meet virtual and cloud performance demands in a smaller footprint and for up to a 30% lower storage equipment costs than using FC drives alone.

–

Provides application-specific controls and comprehensive support for thin and fat volumes, volume copies, and metadata to enable service-level optimization for broad deployments.

Autonomic solution enhances infrastructure agility: –

Enables IT managers to react swiftly to changing business needs and delivers service-level optimization over the entire application life cycle, autonomically and nondisruptively.

–

Offers QoS gradients for application prioritization modes (performance, balanced, cost) to shift data at a granular level toward the most appropriate resources according to changing QoS demands.

–

Intelligently monitors subvolume performance and then applies user-specified policies to autonomically and nondisruptively move data.

Flexible, proven implementation minimizes risk: –

Leverages the fine-grained data movement engine built into HPE 3PAR OS. This data movement engine provides nondisruptive rebalancing of storage volumes using Dynamic Optimization.

–

Protects against user impact by incorporating a high degree of user control and several policy override mechanisms. A flexible implementation allows data movement to be scheduled and resource usage limits and tier definitions to be varied by application.

–

Supports multiple Adaptive Optimization configurations and the coexistence of tiered and nontiered application volumes, delivering the ability to consolidate a wide range of applications onto a single array.

||||||||||||||||||||

Performance benefits of Adaptive Optimization Figure 2-76 shows just an example of how input/output operations per second (IOPS) and latency for a two-tier configuration within a storage system and the combined performance appear under various workloads. The Tier 0 only performance data

Technet24

||||||||||||||||||||

||||||||||||||||||||

typically sets the upper limit, and Tier 1 only performance data typically sets the lower limit. Combining Tier 0 (SSD) and Tier 1 (hard disk data [HDD]) typically has a minimal impact on IOPS but a significant positive impact on latency.

Figure 2-76 Performance benefits of Adaptive Optimization The combined result is dramatic—total IOPS and latency remain close to Tier 0 performance for 8 K random read and 8 K random online transaction processing (OLTP) tests. In the case of 8 K random write tests, the slower numbers reflect the use of Network RAID 10, which is a technology that replicates writes in the storage cluster for high availability. The duplication of writes involving spinning media impacts overall write performance. These numbers demonstrate a best-case example of how Adaptive Optimization can accelerate performance when Tier 0 is 100% used. The degree to which Tier 0 is used in specific cases will depend on the degree to which applications access particular areas of data locality within larger data sets. With a combination of SSD and SAS drives, Adaptive Optimization performance can approach that of an all-SSD system in terms of both IOPS and latency. The system also benefits from SAS drive capacity, which lowers the cost considerably.

HPE 3PAR Priority Optimization (QoS) HPE 3PAR Priority Optimization software for HPE 3PAR StoreServ Storage systems implements and manages a priority policy per Virtual Volume Set (VVset) that serves as a proxy for applications and per Virtual Domains that serves as a proxy for customers or tenants. HPE 3PAR Priority Optimization, configuration shown in Figure 2-77, operates by applying minimum goals for IOPS, latency, and bandwidth where performance for a

||||||||||||||||||||

||||||||||||||||||||

specific application or tenant will not be throttled below these goals. Also, Priority Optimization can apply upper limit control to IO traffic to and from hosts connected to HPE 3PAR StoreServ Storage.

Figure 2-77 HPE 3PAR Priority Optimization (QOS) These minimum goals and limits, called QoS rules, are defined for front-end IOPS and for bandwidth, are applied via HPE 3PAR Autonomic Groups, and are managed via the HPE 3PAR SSMC and the HPE 3PAR CLI. Configure the QoS rules or policies on a Virtual Domain or a VVset, which might contain a single volume or multiple volumes. Enabling a system QoS rule from VVs and VVsets might not be subjected to a named QoS rule. This prevents any bursty or unplanned increase in IO traffic on HPE 3PAR StoreServ from these volumes. In fact, HPE 3PAR Priority Optimization is the industry’s only “Latency goal” feature that enables one, to set up SLAs as low as 500 μs, also making it possible to configure service-level objectives in terms of KB/s and IO bandwidth on a VVset or between different virtual domains HPE 3PAR Priority Optimization is granular, flexible, and easy to configure and monitor, and requires low supervision from storage system administrators. By contrast, some competitive QoS implementations require that you assign a workload to a predefined priority level or that the array gets logically partitioned to reserve part of its system resources to a particular workload. These solutions are neither flexible nor allow real-time enforcement. Features and benefits include: Create and modify threshold limits including IO per second, bandwidth, and latency to meet business needs. Configure thresholds by volume set with real-time enforcement of modified policies within seconds. Set up easily with autonomic volume sets for efficient service levels.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Specify the relative performance of each workload on HPE 3PAR StoreServ. Control and balance the distribution of HPE 3PAR StoreServ disk bandwidth across workloads on a single system. Avoid resource contention in the storage system by throttling the IO of one or more workloads to enable performance of others.

HPE 3PAR Adaptive Flash Cache One of the most difficult workloads for a storage array to handle is a random read workload. A random read workload refers to a sequence of read requests where the sequence of future read requests cannot be predicted based on previous read requests. Random read workloads are independent of the size of the IO request, but it is the generally accepted norm that the IOs in a random read workload are small block (<64 KiB). For random read workloads, assuming the active working set size is large enough compared to the amount of cache on the array, data generally needs to be read from the back-end storage to satisfy the read request. Although a properly sized array assumes no cache hits for random read workloads, in most situations, some limited percentage of the random read workload will be serviced from cache (that is, a block of data that was read previously will be reread by the server before it is flushed from read cache). By increasing the size of the primary read cache on an array or by adding a large second level of read cache (Level 2 [L2] cache), you can increase the probability of a block of random read data to be serviced multiple times out of a much faster cache tier than the spinning media where it resides on the back end of the array. HPE 3PAR AFC, shown in Figure 2-78, is a built-in array functionality of the HPE 3PAR StoreServ that does this by using capacity on SSDs (flash) to act as L2 read cache holding random read data that has been removed from DRAM read cache.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-78 HPE 3PAR Adaptive Flash Cache An additional benefit of AFC is that it might increase the overall IOPS and lower the overall read IO latency an array can deliver by “unloading” a percentage of random read IOPS from spinning media on the back end of the array to much faster flash media. Using SSDs as a Level 2 read cache to hold random read data that has been removed from DRAM cache is a cost-effective way of keeping more random read data on very fast media to improve overall read performance.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Increasing and decreasing the size of AFC The configured Flash Cache size on a system cannot be changed dynamically. To increase or decrease the amount of Flash Cache configured on the system, Flash Cache must first be removed with the removeflachcache command (this will disable Flash Cache for all VVs) and then recreate Flash Cache with the createflashcache command specifying the new desired size. Because Flash Cache must first be removed to change its size, it is recommended that this be done during a maintenance period if possible because it will take some time for the Flash Cache to warm up after being recreated and this might have a noticeable effect on the system performance. All existing configuration rules created with the setflashcache command will persist upon Flash Cache removal and recreation.

Balancing Flash Cache after adding SSDs to the system HPE 3PAR Autonomic Rebalance provided via the tunesys operation does not apply to Flash Cache LDs. After adding additional SSDs to a system, Flash Cache can be rebalanced across the new SSDs by first removing Flash Cache and then recreating Flash Cache. Because Flash Cache must be removed to rebalance across new SSDs is recommended that this be done during a maintenance period if possible because it will take some time for the Flash Cache to warm up after being recreated, and this warm-up period might have a noticeable effect on the system performance. Rules created with the setflashcache command will persist when Flash Cache is recreated.

Limitations AFC operates on small block random read data only (read data with IO size less than 64 KiB). The minimum amount of AFC that can be specified on an HPE 3PAR StoreServ array regardless of model is 64 GB per node pair. The maximum amount of AFC that can be configured per node pair is limited by StoreServ array model. Note For more information, refer to the HPE 3PAR Support Matrix in Single Point of Connectivity Knowledge (SPOCK) on the HPE website. https://h20272.www2.hpe.com/spock/

||||||||||||||||||||

||||||||||||||||||||

Adaptive Flash Cache compared to Adaptive Optimization Even though HPE 3PAR AFC caches small block read data that resides on spinning media onto SSD, it does not compete with nor is it a replacement for Adaptive Optimization (AO). Where AO focuses on moving larger chunks of data (128 MiB “regions”) between different tiers of storage based on historical data access patterns to help save cost by migrating seldom used data to large, slow NL HDDs and more heavily accessed data to a Fast Class (FC) or an SSD tier, AFC focuses on improving small block random read performance for data that resides on spinning media. AFC and AO can coexist effectively on a system. AFC caches small block read data (IO less than 64 KiB) that comes from a spinning media pool (for example, 10 K FC HDDs) into DRAM read cache onto SSD space allocated on the node. Even though both AFC and AO move “hot” data to SSD, the SSD space that is allocated to AFC is not the same thing as an SSD tier found in an AO policy. Although AO moves data between tiers of storage (up to three tiers) in 128 MiB regions, it does so based on how heavily the regions are accessed (region density) over a specified measurement interval and on a scheduled basis. AO also considers both read and write IOs to the regions when making moves. AFC, however, can cache only read data from a node’s DRAM read cache onto AFC SSD space and does so on a realtime basis. Also, although the unit of data that AO operates on is a region (128 MiB), AFC operates on 16 KiB Cache Memory Pages (CMPs). AFC and AO can coexist together and are in fact complementary to one another. The SSDs in an array can be shared between AFC and AO because of the chunklet-level virtualization of SSDs that occurs on the system, or the SSD space allocated to AFC can be dedicated SSDs. This means that you do not have to dedicate specific SSDs as AFC. Instead, you can allocate a certain amount of SSD capacity behind each node to be used as AFC. Read accesses to data that reside in AFC are not included in the region density data collected by HPE 3PAR System Reporter Software. Therefore, they will not have an effect on AO moving the corresponding region to a higher tier as a result of heavy access. This means that it is possible for some applications to have data that, even though it is heavily accessed on a daily basis, does not get migrated to a higher tier of storage by AO. This is because the region accesses are not counted as a result of that hot data ending up in AFC. As a result, if this daily hot data resides on NL HDDs, it might suffer a noticeable performance lag until the data makes its way into AFC.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ Peer Motion and Online Import HPE 3PAR Peer Motion support for nondisruptive, bidirectional data movement allows enterprises to create a storage federation, or an elastic resource pool, across multiple storage arrays with up to 60 PB of aggregate usable capacity, capable of delivering over 10 million IOPS and 300 GB/s bandwidth. With a single click, workloads can move between members of a federation to dynamically rebalance storage resources for cost and performance optimization, rapid response to new business opportunities, and simple technology refreshes. Peer Motion, shown in Figure 2-79, is a feature that allows migration from one HPE 3PAR array (source) to another HPE 3PAR array (target/destination).

Figure 2-79 HPE 3PAR StoreServ Peer Motion and Online Import When migrating from an HPE EVA/P 6000, EMC, or HDS array (source) to an HPE 3PAR array (target/destination), the feature is referred to as Online Import and is only unidirectional. HPE 3PAR Peer Motion—Online Data Migration from HPE 3PAR to HPE 3PAR HPE 3PAR Online Import for EVA—Online Data Migration from EVA Storage to HPE 3PAR StoreServ Storage using an EVA plug-in within the Command View EVA Management interface HPE 3PAR Online Import for EMC Storage—Online Data Migration from EMC Storage to HPE 3PAR StoreServ Storage. Orchestration for HPE 3PAR Online Import via scriptable CLI interface Supports migration from CX4, VNX, VMAX family platforms

||||||||||||||||||||

||||||||||||||||||||

Supports Windows 2008 R2, 2012 (Standalone and Clustered hosts) Supports RHEL 6 Standalone hosts Available for HPE StoreVirtual and HPE 3PAR StoreServ arrays, Peer Motion enables customers to: Ensure constant productivity and redistribute workloads nondisruptively in response to an unpredictable environment with federated workload balancing Increase ROI with federated Thin Provisioning, which transparently moves application data to systems with available capacity, reducing the need for future storage purchases Improve productivity by moving data from retiring arrays to new systems nondisruptively with federated asset management, which eliminates downtime or service interruption during storage asset refresh activities Peer Motion leverages HPE 3PAR Thin Built In technology and HPE 3PAR Thin Conversion software to power the simple and rapid in-line conversion of inefficient, fat volumes on source arrays to more efficient, higher-utilization thin volumes on the destination HPE 3PAR storage system. The destination storage system connects to the source storage system as a peer and imports the data while the host IO continues. In addition to enabling an online migration, support for fat-to-thin conversion during data migration enables cost savings. Zeroes are detected by the Thin Built In ASIC and virtualized by Thin Conversion, thereby removing wasted capacity that was allocated but unused on the source storage system. This allows the purchased storage capacity on the destination storage system to be right sized. Peer Motion Manager orchestrates all stages of the data migration life cycle to ensure that data migration is simple and foolproof. Note Peer Motion is separately orderable software.

HPE 3PAR Persistent technologies The HPE 3PAR architecture features a set of technologies designed to eliminate data corruption, prevent performance degradation during component failure, mask component failure from hosts, and even protect against entire site failure. HPE 3PAR Persistent technologies are shown in Figure 2-80.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 2-80 Persistent technologies Modern Tier 1 resiliency requires that data access and service levels be maintained during failure recovery, maintenance, and software upgrades. Tier 1 resiliency demands that failures not only be prevented, but that the system can recover quickly in the event that something goes wrong. Not only is HPE 3PAR StoreServ Storage designed to be nondisruptively scalable and upgradable, but the system also has several advanced features to prevent unnecessary downtime and to maintain availability and performance levels during planned as well as unplanned outage events. These features are collectively known as persistent technologies.

HPE 3PAR StoreServ Persistent Checksum Persistent Checksum, shown in Figure 2-81, addresses media and transmission errors that can be caused by any component in the IO stack from the server HBA through the SAN switches and into the HPE 3PAR StoreServ HBAs. Persistent Checksum is server and application independent and offers elaborate host operating system support.

Figure 2-81 HPE 3PAR StoreServ Persistent Checksum When supported HBAs are used, T10-DIF2 tags are added and verified on the HPE

||||||||||||||||||||

||||||||||||||||||||

3PAR StoreServ Storage system. When unsupported HBAs are used, T10-DIF tags are added and verified on the array target ports and back-end HBAs, securing the data from the hosts to the drives. Where Persistent Checksum detects media or transmission errors, graceful error recovery will take place, avoiding impact on the host application. Back-end T10-PI (Persistent Checksum) is available and always on for HPE 3PAR 8000, 9000, and 20000 arrays. Array E2E T10-PI is available and always on only for HPE 3PAR StoreServ 9000 and 20000 arrays. Host E2E T10-PI is available for HPE 3PAR StoreServ 20000 and on, if a supported HBA and driver are installed. The initial support is for FC only.

HPE 3PAR Persistent Cache HPE 3PAR Persistent Cache is a resiliency feature built into HPE 3PAR OS that allows graceful handling of an unplanned controller failure or planned maintenance of a controller node. This feature eliminates the substantial performance penalties associated with traditional modular arrays and the cache “write-through” mode they have to enter under certain conditions. HPE 3PAR StoreServ Storage can maintain high and predictable service levels even in the event of a cache or controller node failure by avoiding cache write-through mode via HPE 3PAR Persistent Cache. Under normal operation on an HPE 3PAR StoreServ Storage system, each controller has a partner controller in which the controller pair has ownership of certain LDs. LDs are the second layer of abstraction in the system’s approach to virtualization of physical resources. In the event of controller node failure, HPE 3PAR Persistent Cache preserves write caching by having the surviving node from a node pair, where one node has failed, mirror writes that would have gone to the failed node to other nodes in the cluster instead. By doing this, the surviving node does not have to go into write-through mode for the LDs it owns to ensure data integrity in the unlikely event it should fail too. For example, in a quad controller configuration (where Node 0 and Node 1 form a node pair and Node 2 and Node 3 form a second node pair), each node pair might own 100 LDs with each node in the pair fulfilling the role of the primary node for 50 of those LDs. If Node 2 fails, the system will transfer ownership of its 50 LDs to Node 3, and Node 0 and Node 1 will now be the backup (and thereby the cache mirroring partner) for the 100 LDs that Node 3 is now responsible for. The mirroring of write data coming into Node 3 for those 100 LDs will be distributed across Node 0 and Node 1.

Persistent Ports Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR Persistent Ports enable a nondisruptive environment where host-based multipathing software is not required to maintain server connectivity in the event of a node or link outage on any SAN fabric. This applies to firmware upgrades, node failures, and node ports that are taken offline either administratively or as the result of a hardware failure in the SAN fabric that results in the storage array losing physical connectivity to the fabric. From a host standpoint, connections to StoreServ storage systems continue uninterrupted with all IO being routed through a different port on the StoreServ storage array, as shown in Figure 2-82. This helps you achieve an uninterrupted service level for applications running on HPE 3PAR StoreServ Storage systems.

Figure 2-82 Persistent Ports HPE 3PAR Persistent Port functionality works for FC, FCoE, and iSCSI transport layers and provides transparent and uninterrupted failover in response to the following events: HPE 3PAR OS firmware upgrade HPE 3PAR node maintenance or failure HPE 3PAR array “loss sync” to the FC fabric or iSCSI network Array host ports being taken offline administratively Port laser loss for any reason (applies to FC only)

||||||||||||||||||||

||||||||||||||||||||

Persistent Ports on HPE 3PAR StoreServ systems The addition of virtual ports to HPE 3PAR OS allows 3PAR host-facing ports to mimic the World-Wide Name (WWN) or port of a partner node in the event that the partner node goes down or the link is down, as shown in Figure 2-83. On FC ports, the system uses NPIV. On iSCSI ports, the system uses IP failover. In SAN connect configurations, port switchover keeps host paths online so that online software upgrades can proceed without the risk inherent in any dependency on multipath software.

Figure 2-83 Persistent Ports on HPE 3PAR StoreServ systems Failover features of Persistent Ports include: Failover for FCoE and iSCSI ports—FCoE uses the same underlying NPIV technology Technet24

||||||||||||||||||||

||||||||||||||||||||

as FC. iSCSI uses IP failover. The IP address is transferred and a gratuitous Address Resolution Protocol (ARP) is broadcast. The user experience is the safe as with FC ports. Loss_sync trigger for FC port failover—A failover is triggered if loss_sync is detected by a 3PAR HBA port. There is no failover if the partner port is also in loss_sync. Improved failover speed—Previously, initialization of VLUN data structures did not start until the host logged in to the partner port. With HPE 3PAR OS 3.1.3 and higher, that VLUN initialization is done before failover. Note For more information, access an HPE white https://www.hpe.com/h20195/V2/GetPDF.aspx/4AA4-4545ENW.pdf

paper:

HPE 3PAR Remote Copy software HPE 3PAR Remote Copy software provides enterprise and cloud data centers with autonomic replication and Tier 1 disaster recovery technology that allows the protection and sharing of data from any application simply, efficiently, and affordably. Remote Copy, shown in Figure 2-84, reduces the cost of remote data replication and disaster recovery by leveraging thin copy technology and enabling replication with midrange, all-flash, and high-end arrays to deliver unique cost efficiencies for protecting a broader set of applications.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-84 Remote Copy modes of operation

Remote Copy modes With support for Synchronous, Asynchronous Periodic, multisite, Asynchronous Streaming, and Synchronous Long-Distance replication, Remote Copy delivers a rich portfolio of replication options for customers to achieve stringent RTO and fast RPOs. The unique Synchronous Long-Distance mode allows you to meet low RTO and zerodata loss fast RPOs with complete distance flexibility. HPE 3PAR Utility Storage is the only platform to offer cost-effective, multisite, multimode replication accommodating the use of midrange arrays without requiring expensive service engagements.

Synchronous mode replication Synchronous replication is best suited for mirroring over shorter distances, typically within the same metropolitan area. This mode provides real-time mirroring and ensures

Technet24

||||||||||||||||||||

||||||||||||||||||||

the highest possible data currency by writing data simultaneously to the cache of both systems before the application IO is signalled “complete.” Synchronous mode maintains an up-to-date copy of data that is readily usable at the remote site. Because each IO is mirrored and verified at the remote location before the host application receives an acknowledgement, data consistency of each IO is guaranteed, as shown in Figure 2-85.

Figure 2-85 Synchronous mode replication

Synchronous Long-Distance mode replication Synchronous Long-Distance mode replication combines the complete distance flexibility offered by Asynchronous Periodic mode with the ability to maintain concurrent replicas associated with Synchronous mode replication for rapid application recovery. This is accomplished by using two backup storage servers—one located near the primary site using a Synchronous mode connection (the sync server) and a distant storage server that uses an Asynchronous Periodic mode connection (the disaster recovery server). Synchronous replication keeps source and destination in perfect sync but is at the cost of performance because each write must be written at the source and destination location before being acknowledged. The time taken to acknowledge a synchronous write is minimal, but with the ultra-low latency of an all-flash environment, any additional delays become significant. Asynchronous Streaming replication is a form of asynchronous replication offering a very low RPO measured in seconds. This mode offers a balance of data protection with an RPO of mere seconds but without the

||||||||||||||||||||

||||||||||||||||||||

performance impact or distance limitations of Synchronous replication. This feature is available across all current models.

Asynchronous Periodic mode replication Asynchronous Periodic mode replication offers optimal disk and bandwidth utilization and provides cost-effective replication over any distance. This mode uses nondisruptive, point-in-time synchronization to capture and mirror to the remote site only the data that has changed since the last update. The system does not physically duplicate the data at the time the snapshot is taken but simply maps the snapshot space to the primary volume using a mapping structure consuming virtually no disk space. When the blocks become modified, only the modified blocks consume disk space. For each scheduled Asynchronous Periodic snapshot, only this incremental change is captured and copied to the remote site, enabling quick resynchronizations and low network bandwidth consumption each time, as shown in Figure 2-86.

Figure 2-86 Asynchronous Periodic mode replication

Technet24

||||||||||||||||||||

||||||||||||||||||||

Asynchronous Periodic replication gives you the granularity to control when and how often point-in-time updates occur. This allows you to choose the optimal time for resynchronizations, such as more frequently during critical business hours, or perhaps only at off-peak times when bandwidth is less expensive. FC over IP (FCIP) is supported only by Asynchronous Periodic mode replication.

Asynchronous Streaming mode replication Just like Synchronous and Periodic Asynchronous modes, Asynchronous Streaming Remote Copy ensures IO consistency across the volumes in a Remote Copy group. Asynchronous Streaming mode has been added to the Remote Copy data replication suite. As shown in Figure 2-87, Asynchronous Streaming Remote Copy uses the concepts of replication sets, subsets, and tickets to ensure IO delivery and ordering. Data is held in cache on both the source and target array when it is being replicated. As a result, all solutions must ensure array IO performance is sufficient for the IO load, and there is sufficient replication bandwidth between arrays to prevent cache on either array from filling up with replication data.

Figure 2-87 Asynchronous Streaming mode replication Either having too small of a link or an undersized array can result in Remote Copy groups being suspended. The source array can (and will) use up to 20% of the cache on

||||||||||||||||||||

||||||||||||||||||||

the array for Asynchronous Streaming Remote Copy use. Today, there are no hard limits on how much cache is used on the target array to hold target data. Note For more information on the features and functions of configuring disastertolerant solutions with HPE 3PAR Remote Copy, read the white paper. https://www.hpe.com/h20195/v2/GetPDF.aspx/4AA3-8318ENW.pdf

Replication sets Replication sets provide the granularity on which IO consistency is maintained for a given Remote Copy group. A given replication set is associated with at most one Remote Copy group. Replication sets represent a point in time (100 ms by default). IOs received on the source array during this time will be associated with the same set. A new set is started for every Remote Copy group approximately every 100 ms (subject to change) and is facilitated by incrementing a cluster-wide sequence number. The number of IOs in a set is variable and depends on the number of writes the server made to the VVs in the Remote Copy group over a 100-ms period. A replication set is made up of subsets containing Remote Copy replication tickets for the VVs in a Remote Copy group. If a node on the source array receives a write IO, the data for the write is mirrored to cache on the appropriate node in the cluster and the IO is passed to the Remote Copy layer for replication processing. A replication ticket for that IO is then generated and the IO is added to the current active subset on the node that owns the logical block address for the IO request. Replication tickets are unique across the array and are used to enforce IO ordering and ensure IO delivery. Information that makes up the subset is mirrored to the node’s partner node (policy memory, not data cache). The subset information from a node is mirrored to its partner so that the partner can recover the subset in the event the node fails. The server is then sent an IO complete message and the IO is queued up to be replicated to the remote array. A set is considered complete when a system-wide cluster sequence number increments. When the cluster-wide sequence number increments the current set for all Asynchronous Streaming, Remote Copy groups are complete. After all IOs associated with a given set have been sent to the target array, the source array informs the target that the current set is complete by sending it a manifest record for that Remote Copy set. There is a manifest record for every Remote Copy group that specifies what IOs (replication tickets) and how many IOs are in each subset. This is done in a manner that provides smooth streaming of data between the source and target arrays for a properly sized

Technet24

||||||||||||||||||||

||||||||||||||||||||

solution. When the target array receives the manifest record indicating that the current replication set is complete, it checks to ensure that it has received all IOs that belong in that set. If the target determines that any IOs are missing, it will ask the source array to resend the missing IOs. After the target has all the IOs for the set, it passes the IOs to the VV layer in an IO-consistent manner. The IOs for a set are not applied to the target VV until the set is complete, and all IOs belonging to the set have been received and properly ordered. The Remote Copy tickets are used by the target array to enforce IO ordering and to ensure that all IOs specified in the manifest message have been received.

HPE 3PAR Storage Federation Storage federation, shown in Figure 2-88, is a native functionality built into HPE 3PAR OS that enables users to move data and workloads between arrays without impacting applications, users, or services. Simply and nondisruptively shift data between any model of StoreServ storage system without additional overhead impact to host management layers or appliances. With a very broad support matrix of host operating systems, seamless bidirectional data mobility on StoreServ storage systems can also improve availability in not just physical hosted environments but also hypervisor-based environments such as Hyper-V and VMware vSphere.

Figure 2-88 HPE 3PAR Storage Federation benefits One of the benefits of using HPE 3PAR Federation is that it allows the user to move

||||||||||||||||||||

||||||||||||||||||||

application workloads nondisruptively between storage systems to optimize resource utilization, as well as avoid hotspots and bottlenecks. Storage federation is the delivery of distributed volume management across selfgoverning, homogenous peer storage arrays. Storage federation on StoreServ storage systems is a peer-to-peer relationship, something that is native to the system itself. Today, the maximum number of members in a federation is 4, as shown in Figure 2-89.

Figure 2-89 HPE 3PAR Storage Federation with 4 members The primary use case for multiarray bidirectional data mobility is to be able to move data from one array to another array, having simultaneous migrations going within a federation. Creating a Federation is dependent on two factors: Number of arrays to be included in the Federation: Currently, Federation supports up to four HPE 3PAR arrays in a loosely coupled formation and supports bidirectional data movement between the arrays. Other configurations are supported but not defined in the scope of this document. Zoning: It is the key component in creating a Federation. Each federated member must contain peer ports and host ports, which are zoned to two separate fabrics. Federation is accomplished by connecting node pairs (for example, Nodes 0 and 1) via peer links. Peer links are connected through redundant SAN switches, which serve as a dedicated path for data transfer between the source and destination array.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Note Storage federation is different from hierarchical virtualization, which is the delivery of consolidated or distributed volume management through appliances that hierarchically control a set of heterogeneous storage arrays. Hierarchical virtualization, also sometimes referred to as external storage virtualization, adds a new layer that has to be purchased and managed. A new layer not only introduces additional fault domains that need to be handled but also compromises the overall functionality of the entire system to be at the lowest common denominator. By contrast, storage federation on HPE 3PAR StoreServ Storage systems delivers the following benefits: Keeps costs low (no redundant layer of intelligent controllers) Reduces failure domains (no additional layers) Maintains functionality of each of the peers Simplifies administration and reduced complexity of interoperability between storage systems Storage federation has emerged as a way to address and improve storage agility and efficiency at the data center (and even metropolitan area level). The primary building blocks of HPE 3PAR Storage Federation are HPE 3PAR Peer Motion, HPE 3PAR Online Import, and HPE 3PAR Peer Persistence.

Virtualization compared with storage federation Competitors offer appliances to virtualize and pool storage across storage arrays. With this type of storage virtualization, consolidated or distributed volume management is delivered through appliances that hierarchically control a set of heterogeneous storage arrays. With federation, a set of self-governing, homogeneous, peer storage arrays deliver distributed volume management. The table in Figure 2-90 describes the advantages and disadvantages of both modes.

||||||||||||||||||||

||||||||||||||||||||

Figure 2-90 Advantages and disadvantages of virtualization and storage federation

Technet24

||||||||||||||||||||

Learning check What are primary features of HPE 3PAR OS software? (Select three.) A. Workloads can move between members of a federation to rebalance storage resources dynamically. B. Autonomic provisioning features eliminate traditional storage planning. C. Advanced virtualization capabilities deliver high and predictable service levels. D. Data protection and copy space reclamation features improve capacity utilization. E. Streaming offers optimal disk and bandwidth utilization and provides cost-effective replication over any distance.

||||||||||||||||||||

Name two management tools used to configure HPE 3PAR StoreServ.

HPE InfoSight is a Reactive Analytic tool. True False Name two product families in the HPE primary storage portfolio.

Your customer has a 3PAR StoreServ 8000 and wants to transition to the new allinclusive license. What minimum OS version should they be operating on?

Write a summary of the key concepts presented in this chapter.

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Summary HPE provides a comprehensive portfolio of block-, file-, and object-level storage. Management tools enhance visibility and assist in maintaining updates. HPE provides simplified all-inclusive licensing along with O/S and bundled software titles. Nondisruptive Federation improves efficiencies for migration and lifecycle management.

||||||||||||||||||||

||||||||||||||||||||

3

HPE Virtualization, Fabrics, and Converged Management

Technet24

||||||||||||||||||||

||||||||||||||||||||

OBJECTIVES After completing this chapter, you should be able to: Discuss the fundamentals of virtualization. Explain the benefits of Hewlett Packard Enterprise (HPE) Virtual Connect for BladeSystems. Describe storage area network (SAN) fabric topologies. Identify the virtualization management capabilities provided by HPE OneView. Describe HPE Hyper Converged systems and the benefits they provide. Identify HPE Composable Infrastructure.

||||||||||||||||||||

||||||||||||||||||||

Prelearning check Describe software-defined storage (SDS).

Technet24

||||||||||||||||||||

||||||||||||||||||||

Fundamentals of virtualization Virtualization is the pooling of multiple resources into what appears to be a single device that is managed from a central console. Virtualization: Aggregates physical resources into a unified structure or pool. Presents those resources as capabilities that can be consumed by applications or other types of storage clients. Abstracts storage at the block level. Virtualization improves server utilization, but it also depends on shared storage to meet application requirements more effectively. Virtual environments require high levels of data availability and storage utilization as well as the ability to scale nondisruptively as business needs expand. This gives clear insight into the real costs of storage. At the same time, virtual environments: Reduce costs for multisite, high availability storage by approximately 50%. Increase capacity utilization by approximately 33% to improve storage investments. Deploy shared storage without disrupting business operations. Virtualization exposes storage challenges in the areas of efficiency, availability, and management.

Journey of virtualization Figure 3-1 provides an overview of virtualization continuum.

||||||||||||||||||||

||||||||||||||||||||

Figure 3-1 Journey of virtualization The flexibility of server utilization vastly improved with the introduction of Hypervisors and converged system adding faster time to value which were factory racked and stacked. As the industry moves into hyperconverged and software-defined architectures, virtualization solutions in a single platform are simplifying management again with Factory-integrated systems. The Cloud increased business agility, improving the speed of provisioning with new governance controls and self-service portals. Composable Infrastructure is able to support the Idea Economy for both virtualized and legacy workloads from a single platform offering extreme elasticity, improved infrastructure management and providing rapid orchestration.

Storage integration into hypervisors A hypervisor virtualizes hardware to provide an environment in which you can run multiple operating systems at the same time on one physical server. The hypervisor enables you to create and manage virtual machines (VMs) and their resources. Each VM is an isolated, virtualized computer system that can run its own operating system. The operating system that runs within a VM is called a guest operating system. VM storage is virtualized in a data center that implements server virtualization because all VMs store at least one virtual disk file and a configuration file. A storage hypervisor extends that basic virtualization by managing different types of infrastructures. Technet24

||||||||||||||||||||

||||||||||||||||||||

A storage hypervisor aggregates unused capacity from various physical storage devices into one resource pool. Because it is software based, a storage hypervisor can run on dedicated storage, in a VM, or inside a server hypervisor. When used with server virtualization, a storage hypervisor provides better storage performance. Storage systems designed for VMs can enhance data migration, performance, and integration with virtual server environments. HPE provides management integration for VMware® vCenter™ and Microsoft System Center to allow typical management operations to be performed directly from the hypervisor management software. Storage capacity monitoring and storage provisioning are accessible for the virtualization administrators through HPE OneView for vCenter and HPE OneView for System Center. These components connect the hypervisor management software and the storage systems (all primary storage array families in the HPE portfolio). HPE OneView for vCenter is provided as a virtual appliance, and HPE OneView for System Center is software that is installed on the System Center server(s). HPE recommends these management integrations for all HPE StoreVirtual virtual storage appliance (VSA) installations.

Software-defined storage SDS moves data services, previously provided by external storage arrays, into the server with maintained full control over the storage layer. HPE StoreVirtual VSA, shown in Figure 3-2 and Figure Figure 3-3, enables customers to run enterprise-class storage features on the same set of hardware that also runs the application workload, whether it has virtualized databases or file services.

||||||||||||||||||||

||||||||||||||||||||

Figure 3-2 Software-defined storage

Figure 3-3 HPE StoreVirtual VSA Traditional architectures deploy, compute and storage resources separately, SDS converges storage and compute. SDS is fully virtualized with no physical hardware dependency for provisioning, performance, optimization, tenants, and so on. It allows organizations to create an open pool of shared storage capacity from existing standards-based hardware, and then use the unified application program interfaces (APIs) to drive orchestration between storage resources and other aspects of the data center. It provides maximum cost optimization, flexibility, and opportunities for orchestration across software-defined data centers (SDDCs). It delivers advanced data services such as snapshots, thin provisioning, data protection, and multisite disaster recovery. The underlying hardware technology must be open, standards-based hardware such as x86-based server technology. Rich data services are based in software, a key benefit of SDS. These technologies must be held together by one common management interface. For true portability and flexibility, these should be open, API-based management, and orchestration tools. SDS allows business applications and underlying storage services to share hardware resources. By converging applications and storage on the same platform, customers can improve utilization rates of compute power and increase the efficiency of power and cooling resources. SDS features include: Provisioning—There is no need to specify drives to pools for provisioning or optimization. Performance—No dedicated hardware is associated with delivering a given amount of performance to a volume or application. Not segregated—Dedicated hardware is associated with delivering mixed workload Technet24

||||||||||||||||||||

||||||||||||||||||||

performance. Tenants—Secure tenant domains are not associated with dedicated physical resources.

HPE StoreVirtual VSA HPE StoreVirtual VSA is a VSA that creates highly available shared storage from direct-attached storage (DAS) in VMware vSphere, Microsoft Hyper-V, and Linux Kernel-based Virtual Machine (KVM) environments. The software-defined primary storage offering, HPE StoreVirtual VSA, is a VSA that creates highly available shared storage from DAS in VMware vSphere® and Microsoft® Hyper-V environments. Its platform flexibility allows you to create a virtual array within and across any x86 server and nondisruptively scale capacity and performance as workload requirements evolve. The ability to use internal or external storage within your environment greatly increases storage utilization and eliminates the costs and complexity associated with dedicated external SAN storage. Storage pools based on StoreVirtual VSA are optimized for vSphere and Hyper-V and typically scale linearly from 2 up to 16 nodes. Integrated with HPE Recovery Manager Central for seamless data movement to StoreOnce Backup appliances and StoreOnce VSA, and bidirectional replication with 3PAR StoreServ arrays (Peer Copy). HPE StoreVirtual VSA Ready Nodes—HPE tested reference configurations on HPE ProLiant servers that address typical virtualization and remote office use cases, and deployment of converged solutions. Support for VMware vSphere v6.5 and Microsoft Hyper-V Server 2016. Perpetual license support for 10TB and 50TB licenses now available. Expanded open-source virtualization platform support for KVM versions: CentOS, RHEL, and Ubuntu. Stateful containers with persistent storage can be deployed with Docker through either a native Docker plug-in or through Cinder drivers in an OpenStack environment for flexibility and agility. With the wide adoption of virtualization and the evolution of SDS technologies, it is important to understand the decision points within the infrastructure that can directly influence workloads. A key benefit of SDS is the flexibility to build a system to match workload requirements.

||||||||||||||||||||

||||||||||||||||||||

When designing an SDS solution based on StoreVirtual VSA, it is important to understand the performance and capacity requirements of the environment, especially during peak activity. After you understand the storage requirements, you can design a solution with the right disk drives, server platforms, and network infrastructure.

HPE StoreVirtual VSA Ready Nodes One of the benefits of SDS, having the flexibility of the underlying server platform, can also be a challenging aspect with the vast variety of server platforms and server options. By using HPE StoreVirtual VSA Ready Nodes, you can accelerate the overall deployment time and simplify the installation of SDS solutions. Based on the information provided by system architects, you can build converged compute and storage for virtualization projects guided by a configuration tested by HPE. HPE StoreVirtual VSA Ready Nodes are reference configurations to provide customers and partners a starting point when selecting configurations for virtualization projects on the HPE ProLiant server platform. HPE StoreVirtual VSA Ready Nodes will be refreshed over time as new generations of servers, and new options become available. There are four predefined configurations: Small—Small and medium businesses; remote and connected offices Medium—Virtualization in large remote and connected offices Medium Hybrid—Performance server virtualization projects Large Hybrid—High-density server virtualization projects

SDS Blueprints Collectively, these technical briefs provide a basic overview of StoreVirtual technology deployment. SDS blueprints depict typical configurations for HPE solutions, to give you an idea of how to deploy HPE StoreVirtual technology in the field. Each blueprint in the series includes a brief description of customer challenges, an example SDS configuration that addresses those challenges, a simple, high-level diagram, a list of suggested software and hardware components, and recommendations for expanding or modifying the solution for different environments.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Blueprint no. 1 compares traditional data storage with SDS technology. It includes definition of VSA, basic features and benefits, and environments for deployment. An example from blueprint no. 1 is provided in Figure 3-4.

Figure 3-4 Example from SDS blueprint Blueprint no. 2 describes the challenges of traditional IT infrastructures and shows how SDS addresses these challenges. It depicts HPE StoreVirtual VSA installed on one or two individual servers at a single site. Blueprint no. 3 explores the limitations of physical storage systems and storage silos. The technical brief describes StoreVirtual VSA deployed on VMs for shared storage in virtualized server environments and illustrates a storage cluster at a single site. Blueprint no. 4 addresses unpredictable workloads and investment protection in mixedvendor IT environments. It depicts an all-flash, low-cost, high-performance StoreVirtual VSA storage solution at a single site. It includes a description of adaptive optimization, an auto-tiering feature that improves data accessibility. Blueprint no. 5 includes an HPE BladeSystem c7000 enclosure at the data center, and StoreVirtual VSA installed on industry-standard servers at remote offices. BladeSystem is a modular infrastructure platform that converges servers, storage, and networking in virtual- and cloud-computing environments, significantly reducing capital costs and data center footprint. Blueprint no. 6 shows a cost-effective, simple configuration for small- to midsize businesses looking to deploy SDS at remote sites with HPE StoreVirtual 4335 Hybrid

||||||||||||||||||||

||||||||||||||||||||

Storage Systems at the main office. Blueprint no. 7 describes open-platform storage solutions comprised of HPE Hyper Converged appliances and StoreVirtual VSA. The configuration includes StoreVirtual VSA at the data center and HPE Hyper Converged 250 appliances, which can be deployed quickly and managed by IT generalists, in multiple remote sites. Note For more information about planning a storage system using VSA and VSA Ready Nodes and Blueprints, use the following hyperlinks: https://www.hpe.com/h20195/v2/GetPDF.aspx/4AA4-8440ENW.pdf http://h20195.www2.hpe.com/v2/GetDocument.aspx? docname=a00000521ENW https://h20195.www2.hpe.com/V2/GetDocument.aspx?docname=4AA62461ENW

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check Describe software-defined storage.

||||||||||||||||||||

||||||||||||||||||||

VMware Virtual Volumes Virtual Volumes (VVols), shown in Figure 3-5, were introduced with the release of VMware vSphere 6.0. VVols simplify storage operations and management in virtualized environments by: Enabling vSphere to interact with the array. Making manual tasks easier. Opening up communication between vSphere and storage administrators. Allowing provisioning to happen quickly based on policy-driven automation. Supporting policy-based management.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-5 VMware Virtual Volumes VVols—represents significant innovation by changing the way that vSphere VMs interact with shared storage. More specifically, VVols introduce two major changes: VM-level granularity by introducing a one-to-one mapping of VMs to storage volumes, Support for VMware’s storage policy-based management (SPBM) to simplify storage management and automate storage provisioning. Before the introduction of VVols, storage arrays primarily integrated with vSphere at the datastore level using VMware’s VMFS (Virtual Machine File System). Moving forward, users can choose to use VMFS or VVols (or both), with VVols offering more

||||||||||||||||||||

||||||||||||||||||||

advanced capabilities. VMware is continually working to develop and evolve the VVols specification. As a result, certain storage features are not yet supported with VVols. Before implementing VVols, you should consider whether they are a good fit for your environment, based on the current support and limitations in vSphere with VVols. The following vSphere and storage array features are either not supported or have limited support with VVols in vSphere 6.0: Array-based replication: Before vSphere 6.5, any type of array-based replication was not supported at the VM level with VVols. If you need to replicate VMs to another storage array, you should continue doing this at the VMFS level or leverage vSphere 6.5, which supports array-based replication of VVols. vSphere replication: VMware does support using vSphere replication (host based) with VVols, which operates independently of storage array replication at the VM level with VVols. VMware® vCenter™ Site Recovery Manager™: The use of VMware vCenter Site Recovery Manager (SRM) either with array-based or vSphere replication is not supported. vSphere Metro Storage Cluster (vMSC): There is no certification yet from VMware on using VVols in a vMSC configuration. Scalability: As of the HPE 3PAR OS 3.3.1 release up to 2500, VMs are supported using VVols on a single HPE 3PAR StoreServ array. The performance characteristics when using VVols do not change when compared to VMFS because it is mainly dictated by the storage array configuration regardless of the type of vSphere Storage Container you use. You should not base your decision on using VVols, instead of VMFS, based on the expectation that performance will differ. VVols will offer performance comparable to RDMs (raw device mapping). However, while the physical performance characteristics do not change with VVols, the new VM-level granularity that VVols provide will give you finer grain control when applying performance-based storage features to individual VMs. If you are implementing VVols in an existing vSphere environment, you can migrate any existing VMs on VMFS datastores to VVols, assuming the needed capabilities are available with VVols. Currently, there is no method in vSphere to convert entire VMFS datastores into VVols. You can migrate existing VMs on VMFS datastores to VVols using the VMware vSphere® Storage vMotion® feature. To accomplish this, you need ample-free space on your HPE 3PAR StoreServ array to use as VVol storage, which may require adding additional storage or consolidating existing storage to free up space.

Technet24

||||||||||||||||||||

||||||||||||||||||||

After you have sufficient free space available on your HPE 3PAR StoreServ array, you can begin creating new VMs or migrate existing VMs using VVols. In a typical data center, storage arrays have different capabilities that should be aligned with the requirements of individual applications. Without VVol functionality, the storage administrator was responsible for this alignment; storage capabilities could not be applied directly to individual VMs and had to be applied to a datastore/logical unit number (LUN), which often contains numerous VMs. VVol functionality leverages the enhanced vSphere APIs for Storage Awareness (VASA), allowing vSphere administrators to assign storage profiles on a per-VM basis and choose storage system capabilities for every application. HPE supplies storage providers that can integrate with vSphere and support VVols. Note For more information on VVols, refer to the VMware vSphere Virtual Volumes: Getting Started Guide. http://www.vmware.com/files/pdf/products/virtualvolumes/vmw-vspherevirtual-volumes-getting-started-guide.pdf

VVol integration with shared storage The VVol architecture is part of the VMware VASA 2.0 specification, which defines the architecture for VM-level storage array abstraction. VASA 2.0, shown in Figure 3-6, includes interfaces to query storage policies that enable VMware Storage Policy-Based Management (SPBM) to make intelligent decisions about virtual disk placement and compliance.

||||||||||||||||||||

||||||||||||||||||||

Figure 3-6 VASA integration The VASA 2.0 specification describes using VVols to provide access and manageability to each VM data store. Each VMware VM disk is provisioned as a separate VVol within the storage system. A single point of access on the fabric is provisioned through a protocol endpoint from the host to the storage. The endpoint is rediscoverable using regular LUN discovery commands. The VM-level granularity of VVols enables customers to apply the rich array-based data services built into HPE 3PAR StoreServ arrays to specific applications and VMs. This integration enables you to: Maintain efficiency by automatically reclaiming space from deleted or migrated VMs using capabilities built into the HPE 3PAR ASIC. Accelerate performance by automatically detecting and moving the most frequently accessed VMs onto flash-extended cache using HPE 3PAR Adaptive Flash Cache. Eliminate data sprawl by performing the migration functions at the VM or application level with HPE 3PAR federation technologies.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Note For more information on VASA for storage, open the hyperlink. http://blogs.vmware.com/vsphere/2011/08/vsphere-50-storage-features-part10-vasa-vsphere-storage-apis-storage-awareness.html

Hyper-V storage Microsoft Hyper-V enables you to create and manage a virtualized computing environment by using virtualization technology that is built in to Microsoft Windows Server. The Hyper-V infrastructure provides virtualized applications and workloads to support a variety of business goals aimed at improving efficiency and reducing costs. There are several ways to design storage for a Hyper-V environment. It is possible to build a storage solution that uses more than one component, as shown in Figure 3-7. This allows you to use high-performance hardware to satisfy needs of highperformance VMs and allows slower and lower-cost storage equipment to handle the more basic tasks.

Figure 3-7 Hyper-V Storage The Hyper-V technology virtualizes hardware to provide an environment in which multiple operating systems can run at the same time on one physical computer. Hyper-V enables the user to create and manage VMs and their resources. Each VM is an isolated, virtualized computer system that can run its own operating system. The operating system that runs within a VM is called a guest operating system.

||||||||||||||||||||

||||||||||||||||||||

Hyper-V provides the infrastructure for virtualized applications such as Exchange 2013, SQL Server 2014/2016, and SharePoint. Also supported with Hyper-V and HPE Storage: ODX allows Hyper-V and OS to move storage faster and more efficiently, enables protocol and transport agnostic, cross-machine storage subsystem assisted data transfer. Practically eliminates load on the server, enables a significant reduction in the load on the transport network and presents an opportunity for innovation for the storage subsystem. Note Windows Offloaded Data Transfer (ODX, also known as copy offload) enables direct data transfers within or between compatible storage devices without transferring the data through the host computer. For more information about ODX, see the following Microsoft website: http://technet.microsoft.com/en-us/library/hh831628.aspx Used for VM Migration, VM Instantiation, General Bulk Data Movement (such as Large Database Files, HD Video, etc.) Thin Provisioning support, providing mechanisms for identifying thinly provisioned LUNs throughout the OS, ability to query the mapped/unmapped state of LUN extents. Optimization, automatically Reclaim Space with UNMAP, Scheduled UNMAP runs at times of low server IO/CPU utilization and at scheduled times, also runs at time of file deletion. Notification, exposing events to indicate when LUNs cross threshold boundaries. Temporary and permanent resource exhaustion handling.

Comparing VMware and Hyper-V Choosing between VMware and Hyper-V has becoming more difficult because of the fact that both now offer very similar hypervisor technologies. Figure 3-8 displays VMware and Hyper-V comparison from blog by Ken Leoni: https://blog.heroix.com/blog/help-choosing-between-vmware-and-hyper-v

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-8 Comparing VMware and Hyper-V Virtualizing a relatively small number of servers and hosts? Then, Hyper-V may hold a distinct advantage, if you can manage your virtualization with just the free Hyper-V

||||||||||||||||||||

||||||||||||||||||||

Manager. VMware does make a free version of ESXi available; however, in addition to its technical limitations, there is no support. Growing out your environment? Virtualizing with vCenter Server, in general, requires less steps and is quicker than the System Center Suite, resulting in a lower total cost of ownership (TCO).

Desktop virtualization Desktop virtualization is the concept of separating the logical desktop from the physical machine. A virtual desktop infrastructure (VDI) is a form of desktop virtualization that involves more advanced hardware virtualization. Rather than interacting with a host computer directly using a keyboard, mouse, and monitor, the user interacts with the host computer using another desktop computer or a mobile device by means of a network connection, such as a LAN, wireless LAN, or even the Internet. The host computer in this scenario becomes a server capable of hosting multiple VMs at the same time for multiple users. Figure 3-9 provides an example of VDI components.

Figure 3-9 Desktop virtualization

Technet24

||||||||||||||||||||

||||||||||||||||||||

As organizations continue to virtualize and converge their data center environment, client architectures such as VDI also continue to evolve to take advantage of the predictability, continuity, and quality of service delivered by their converged infrastructure. For example, HPE and IBM provide a hybrid VDI model with a range of virtualization software and delivery models to address the limitations of distributed client computing. VDI deployments are great candidates for Data Deduplication because the virtual hard disks that drive the remote desktops for users are essentially identical. Additionally, Data Deduplication can help with the so-called VDI boot storm, which is the drop in storage performance when many users simultaneously sign in to their desktops to start the day. Note VDI is a prime application for both Hyper-V Server and deduplication because the server component does not need licensing and VDI can often achieve significant savings from deduplication. VDI is also the primary solution to use when updating deduplication for virtualization loads. When implementing VDI and Microsoft deduplication, consider using a file server on Windows Server 2016 to store them. Another form of desktop virtualization is called session virtualization. This allows multiple users to connect and log in to a shared but powerful computer over the network and use it simultaneously. Each user is given a desktop and personal folder to store their files. With multiseat configuration, session virtualization can be accomplished using a single PC with multiple monitors, keyboards, and mice connected. Certain client environments move workloads from PCs and other devices to data center servers, creating well-managed virtual clients, with applications and operating environments hosted on servers and storage in the data center. This means that users can access their desktop from any location without being tied to a single client device. Client virtualization enables IT administrators to provide a consistent, seamless desktop experience for users who conduct business transactions on multiple device types, including mobile devices, tablets, laptops, and desktops. IT administrators can use a more centralized, efficient client environment that is easier to maintain and can quickly respond to the changing needs of users and the business. HPE offers all the components needed for a client virtualization implementation: software, hardware, and services, along with a proven methodology to enable faster installation of the right solution the first time. An HPE Converged Infrastructure views

||||||||||||||||||||

||||||||||||||||||||

server, storage, and network resources as pools that are allocated as needed to optimize the delivery of business services while simplifying IT operations. With an HPE Converged Infrastructure, IT organizations can overcome many of the common challenges associated with operating a virtualized client environment. The standardized infrastructure in an HPE Converged Infrastructure offers application mobility, IT simplicity, and lower costs. The ability to migrate applications and workloads efficiently across tightly integrated network, storage, and compute resources can greatly enhance both IT and business operations continuity.

Who uses VDI? There are numerous benefits from using VDI; the percentage of users depend often on the applications, as shown in Figure 3-10.

Figure 3-10 Use cases of VDI VDI infrastructure allows users to access a device based on a Windows operating system, leave for the evening, and come back the next workday to find all data and customized desktop and applications settings intact. A standard VDI configuration might use rack-based servers distributed across the data center with top-of-rack (ToR) switches at the network edge or HPE BladeSystem deployment.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE Nimble solutions for virtual desktop infrastructure HPE Nimble solutions offer: All-flash for absolute performance, Adaptive flash for cost-effective performance. Handle Boot and AV storms with All-Flash performance of 1.4M read, 1.2M write IOPS (input/output operations per second). Move workloads between all-flash and adaptive flash without impacting app layer. Take the guesswork out of scaling VDI. Dedupe and compression ensures capacity savings, enables ‘right-sized’ deployment. Scale from small to large deployments seamlessly and nondisruptively. HPE InfoSight accurately forecasts future capacity, performance, and bandwidth needs. Simplify VDI management. Whether VDI data is on all-flash or adaptive flash arrays, all are managed as one. Figure 3-11 provides an example of the Nimble Storage used for VDI.

Figure 3-11 HPE Nimble and VDI

||||||||||||||||||||

||||||||||||||||||||

HPE InfoSight VMVision HPE InfoSight VMVision™ agentless per-VM monitoring feature brings enterprise IT staff clear visibility into latency and performance across host, network, and storage layers of the stack through intuitive graphical representations. Figure 3-12 and Figure 313 shows the vCenter configuration in Nimble management.

Figure 3-12 VMware integration

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-13 VMware integration The cloud-connected management engine delivers complete visibility through the entire stack and captures latency problems in intuitive pictures that enable IT teams to quickly make highly informed decisions. Its granular insights into each VM’s performance and its unprecedented clarity in presenting potential performance problems help you rapidly pinpoint and resolve issues. When you want to see your way clear in a VM environment to maintain service-level agreements (SLA) and reduce the risks of performance issues, HPE InfoSight VMVision is the intelligent tool that offers you the expertise to do the job quickly and at no additional cost. HPE Nimble InfoSight VMVision clearly shows latency and performance across the storage, network, and host layers through highly intuitive graphics. VMVision helps synergize storage, virtualization, and application teams by mapping out latency at each layer and pinpointing rogue and “noisy neighbor” VMs in a datastore before they cause issues. VM Vision extends the capabilities of InfoSight beyond storage, allowing the customer to opt-in (or out, if they so choose) to send performance data from vCenter. When enabled, we query vCenter every 10 minutes or so and send the payload to InfoSight. Customers can log into the Nimble portal and view performance data in an intuitive way. Features include: Customers opt In Must have vCenter Plugin Registered Minimum NimbleOS 2.2.6 Performance data gathered from vCenter and sent to InfoSight by the array Multiple Dashboards Views available Host Activity Top VMs Datastore Treemap

||||||||||||||||||||

||||||||||||||||||||

VMVision Host Activity dashboard, shown in Figure 3-14, gives an overview of host performance over the past several hours.

Figure 3-14 Host Activity dashboard

Top VMs by IOPS and latency Top VMs dashboard, shown in Figure 3-15, shows the heavy hitters. Great place to look for outliers in the environment. Shows top VMs by I/O and top VMs by latency. Period is last 24 hours. Inactive VMs—VMs that have not generated any I/O over the past 7 days are listed.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-15 Top VMs by IOPS and latency

Datastore TreeMap Datastore Treemap, shown in Figure 3-16, is a visualization tool to show the IO and latency of datastores and VMs.

Figure 3-16 Datastore TreeMap Looking at this view, we see datastores. The larger the rectangle, the more I/O over the

||||||||||||||||||||

||||||||||||||||||||

last 24 hours. The darker the color, the higher the latency. If we then click on a particular datastore, it will give us the list of VMs within that datastore.

Latency analysis Note in Figure 3-17 that we can break down VM latency by: Host (blue) Network (gray) Storage (green)

Figure 3-17 Virtual Machine Latency VM Vision allows us to correlate latency at the VM with all of the factors listed in the slide, for further help in deducing the cause of the latency.

Virtual machine view Figure 3-18 shows an example of comparing an individual VM against a host performance.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-18 Virtual machine view We can also compare individual VM level latencies against the datastore performance or individual virtual disks. We can even zoom into a particular time to compare.

Operational Dashboard VM Vision is so important; we put the top VMs dashboard right on the front page of InfoSight, as shown in Figure 3-19.

||||||||||||||||||||

||||||||||||||||||||

Figure 3-19 Operational Dashboard in InfoSight

VMware Horizon conceptual architecture HPE Client Virtualization is an end-to-end client virtualization solution based on the best managed and virtualization-ready HPE ProLiant and BladeSystem servers, HPE storage, and HPE Thin Client access devices, as shown in Figure 3-20. It is a desktop replacement solution that delivers to IT teams the flexibility to quickly deploy and refresh desktops, the reduction of associated potential data loss or theft. It also lowers the complexity and cost of desktop management, while continuing to provide end users with the functionality of a standalone desktop.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-20 Components of VDI solution VMware Horizon 7 delivers virtualized or remote desktops and applications through a single platform to end users. These desktop and application services—including RDShosted apps (Remote Desktop Services), packaged apps with VMware ThinApp, and Software as a Service (SaaS) apps—can all be accessed from one unified workspace across devices, locations, media, and connections. Note An RDS host is a server computer that hosts applications and desktop sessions for remote access. An RDS host can be a VM or a physical server. An RDS host has the Microsoft Remote Desktop Services role, the Microsoft Remote Desktop Session Host service, and Horizon Agent installed. Remote Desktop Services was previously known as Terminal Services. The Remote Desktop Session Host service allows a server to host applications and remote desktop sessions. With Horizon Agent installed on an RDS host, users can connect to applications and desktop sessions by using the display protocol PCoIP or Blast Extreme. Both protocols provide an optimized user experience for the delivery of remote content, including images, audio, and video (VMware Docs, April 2018). Leveraging closed loop management and optimized for the SDDC, Horizon 7 helps IT control, manage, and protect all of the Windows resources end users want, at the speed they expect, with the efficiency business demands. We have always supported home directories and user shares located on physical

||||||||||||||||||||

||||||||||||||||||||

desktops, and we now also support placing user data on file storage with VMware Horizon VDI desktops. The master images of persistent desktops would still be stored on block, while the user home directories and file shares are stored on file shares on 3PAR. This would allow for QoS (Quality of Service) where the OS images are stored on the solid-state drive (SSD) tier with the file data on spinning disk. Introduction of Horizon Standard, Standard Add-on, Advanced, and Enterprise editions with 1yr, 3yr, and 5yr support, physical format. Introduction of Horizon Standard to Advanced Upgrade for Concurrent Users, Horizon Advanced to Enterprise Upgrade for Concurrent Users, and Horizon Advanced to Horizon Enterprise for Named Users with 1yr, 3yr, and 5yr support, electronic format.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check What information is visualized in the TreeMap tool? What needs to be registered to provide vCenter credentials?

||||||||||||||||||||

Activity: Implementing VMware VVs on HPE 3PAR StoreServ Instructions: Use the “Implementing VMware Virtual Volumes on HPE 3PAR StoreServ” documentation at: https://h20195.www2.hpe.com/v2/getpdf.aspx/4AA5-6907ENW.pdf for this activity. Answer the following questions: –

What is a protocol endpoint?

–

What is VMware SPBM?

–

Who provides the “Storage Capability”?

–

What defines the QoS?

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

HPE Virtual Connect FlexFabric After the introduction of hypervisors more than a decade ago, VM density and mobility increased, as a result of more powerful processor and memory subsystems. This caused greater performance demands on the network subsystems at the server-network edge. HPE Virtual Connect for BladeSystem simplifies server and VM connections to LANs and SANs, vastly freeing up resources of time and power. A Virtual Connect solution is a hardware-based solution that simplifies network I/O by splitting server network connection into four variable partitions. Virtual Connect also simplifies storage, server, and network management in virtualized environments. It frees administrators from the constraints of a traditional infrastructure by bringing “wire-once” simplified management between networks and servers. Wire-once technology can help administrators add, move, or change servers in minutes, thus saving time and improving agility—and eventually providing direct connections to thousands of servers. It also shrinks functional silos between the server, storage, and network environments by giving administrators the tools they need to collaborate in the management of a virtualized data center. In addition, the HPE Virtual Connect FlexFabric architecture improves network utilization and performance. Ports overview shown in Figure 3-21:

Figure 3-21 HPE Virtual Connect FlexFabric 20/40 F8 for c-class BladeSystem 1. Module status, LEDs 2. Module locator (Unit identification, UID)

Technet24

||||||||||||||||||||

||||||||||||||||||||

3. Q1-Q4 QSFP+ 40Gb ports 12. X1-X4 Flexports 13. X7-X8 paired Flexports 14. X5-X6 paired Flexports 4-11 and 15-22 are also status LEDs The HPE Virtual Connect FlexFabric-20/40 F8 Modules, shown in Figure 3-21, are the simplest, most flexible way to connect virtualized server blades to data or storage networks. VC FlexFabric-20/40 F8 modules eliminate up to 95% of network sprawl at the server edge with one device that converges traffic inside enclosures and directly connects to external LANs and SANs. Using Flex-10 and Flex-20 technology with Fibre Channel over Ethernet (FCoE) and accelerated integrated Small Computer System Interface (iSCSI), these modules converge traffic over high-speed 10Gb/20Gb connections to servers with HPE FlexFabric Adapters. Each redundant pair of Virtual Connect FlexFabric modules provide eight adjustable downlink connections (six Ethernet and two Fibre Channel [FC], or six Ethernet and two iSCSI or eight Ethernet) to dual-port 10Gb/20Gb FlexFabric Adapters on each server. Up to twelve uplinks with eight Flexport and four QSFP+ interfaces, without splitter cables, are available for connection to upstream Ethernet and FC switches. Including splitter cables up to 24 uplinks are available for connection to upstream Ethernet and FC. VC FlexFabric-20/40 F8 modules avoid the confusion of traditional and other converged network solutions by eliminating the need for multiple Ethernet and FC switches, extension modules, cables, and software licenses.

FlexFabric and HPE 3PAR StoreServ It is essential to be able to consolidate thousands of VMs onto a single storage system that is connected via Virtual Connect. With highly scalable HPE 3PAR StoreServ, customers get this benefit in addition to greater performance and nondisruptive workload migration. Virtual Connect attaches to HPE 3PAR StoreServ through an intermediate SAN fabric to create a storage solution. However, this fabric can be expensive, and it can result in increased complexity and IT infrastructure costs. With HPE Virtual Connect DirectAttach FC for HPE 3PAR StoreServ, the need for an expensive intermediate SAN fabric

||||||||||||||||||||

linking Virtual Connect and HPE 3PAR StoreServ devices no longer exists. HPE Virtual Connect Direct-Attach FC for HPE 3PAR StoreServ creates a flat SAN architecture, which simplifies problems arising from storage solution complexity. The direct-attach FC solution combines the scalability of HPE 3PAR StoreServ with the simplicity of Virtual Connect. In addition to being much more cost efficient, it provides easy management of the storage solution and frees up valuable IT resources. HPE Virtual Connect Direct-Attach FC for HPE 3PAR StoreServ helps businesses: Reduce costs –

Eliminate the need for expensive SAN fabrics, host bus adapters (HBAs), and cables.

–

Save on operating costs and cut down on capital expenditures.

–

Scale with the “pay-as-you-grow” model, which lets you pay for only what you need now.

Overcome complexity –

Virtual Connect FlexFabric FC connects directly to HPE 3PAR FC storage.

–

Improve efficiency with automated fabric management including simplified management tools.

–

Configure Virtual Connect as direct attach or fabric attach, depending on the solution design.

Simplify management –

Manage through a single pane of glass with HPE OneView and a command line interface.

–

Use Virtual Connect technology to further improve management efficiency.

–

Reduce disparity with separate fabric and device management.

||||||||||||||||||||

HPE flat SAN architecture Flat SAN, shown in Figure 3-22, is a great option for any customer who has HPE blade systems. It basically gives customers all the benefits of a SAN, without the cost and added latency of external switches because the FlexFabric modules are built right into the blade system. When creating a configuration with a flat SAN, some things to pay attention to are:

Technet24

||||||||||||||||||||

||||||||||||||||||||

It is supported with FC, iSCSI but not FCoE host connectivity. For FC, there are no 16Gb Flex Fabric options. So even though the 3PAR FC ports are capable of 16Gb, they will be limited to 8Gb to those hosts. You can mix connectivity types from a single 3PAR array—so you can connect some 3PAR ports to FlatSAN and other ports to a SAN switch. Make sure that the blade servers have been configured with the FlexFabric modules.

Figure 3-22 HPE flat SAN architecture

||||||||||||||||||||

||||||||||||||||||||

For more details about FlatSAN, you can read the HPE Virtual Connect FlexFabric cookbook, and for support details, consult HPE SPOCK.

Convergence with Virtual Connect FlexFabric modules and adapters Legacy Ethernet has been unable to support the requirements of lossless, block-based storage traffic. Data center bridging (DCB), also called converged enhanced Ethernet, allows lossless behavior even in heavily congested networks. This makes it ideal for handling block storage traffic in a converged network environment. FCoE protocols perform best on DCB-enabled Ethernet networks so that Ethernet can carry storage traffic. FlexFabric adapters use DCB-standard packets to encapsulate FC as FCoE, as shown in Figure 3-23. The IEEE 802.1 Working Group is developing these standards for DCB. However, because protocols for converged networks throughout the data center are not yet final, HPE uses known technology requirements to converge networks only at the server-network edge in BladeSystem enclosures.

Figure 3-23 Virtual Connect FlexFabric convergence Virtual Connect technology splits the converged stream to native Ethernet LANs and FC SANs at the FlexFabric modules before going to external connections. HPE views this approach as the best opportunity to take advantage of converged network benefits, where the standards are the most stable, for the greatest return on investment and reduced total cost of ownership.

Technet24

||||||||||||||||||||

||||||||||||||||||||

The FlexFabric adapter encapsulates FC packets as FCoE and consolidates the FC and IP traffic into one 10Gb/s data stream. The FlexFabric interconnect module separates the converged traffic. FC and IP traffic continue beyond the server-network edge using the existing native Ethernet and FC infrastructure. FlexFabric Switches: Support multiple storage networking protocols—FCoE, FC, iSCSI. Can be used as leaf or spine switches. When configured in a leaf-spine fabric topology, a leaf or ToR switch typically connects to servers and storage; a spine switch typically connects to other switches in the network. In converged-Ethernet network terminology, single-layer switch topologies are often referred as a leaf or ToR network topologies. Multihop switch topologies are often referred as a leaf-spine network topology. Support multiswitch networks with up to 30 switches and 7 hops (FCoE) or 3 hops (FC). Convergence—Many FlexFabric switches have 10GbE, 40GbE, or converged ports (10GbE/4/8Gb FC). The 10GbE and 40GbE ports can be used as Ethernet data ports, Intelligent Resilient Framework (IRF) ports, or FCoE ports. The converged ports can be used as Ethernet data ports, IRF ports, or 4 or 8Gb native FC ports.

HPE FlexFabric 5900CP and HPE FlexFabric 5930CP switches The FlexFabric 5900CP and 5930CP switches are ToR converged Layer 2/Layer 3 switches with 48 converged ports and four 40GbE QSFP+ ports. The switches provide seamless integration of 10GbE networks into existing server-edge environments (FCoE DCB, FC) and the HPE server and storage ecosystem. These switches are fully featured, and no software licensing is required. Switches can be added in a FlexFabric 5900CP series IRF fabric. An IRF fabric appears as one node and is accessible at a single IP address on the network. You can use this IP address to log in at any member device to manage all members of the IRF fabric. Note For more information about IRF usage in a storage configuration, see the HPE FlexFabric 5900CP Solutions Configuration Guide. http://h20565.www2.hpe.com/hpsc/doc/public/display? sp4ts.oid=5221896&docLocale=en_US&docId=emr_na-c04216045

||||||||||||||||||||

||||||||||||||||||||

Considerations when designing a FlexFabric switch environment When you are designing a storage solution, you need to consider the infrastructure, relevant storage components, and steps to move the design forward. Specific factors need to be considered at the customer site before implementation. When designing a FlexFabric switch environment, answer these questions about the customer environment before designing the solution: Is there Ethernet/FCoE support on 10GbE converged ports or 40GbE ports? Is there FC 8Gb, 4Gb, or 2Gb support on converged ports? Should the environment support DCB or FCoE-FCF? Is there FCoE or FC NPV gateway support? Does the customer have dual-hop support with HPE Virtual Connect blade switches and the HPE 6125XLG Ethernet Blade Switch? Does the customer have multihop support using FCoE VE_Port (seven hops) or FC E_Port (three hops) ISLs (inter-switch links)? Does the environment have FlexFabric 5900CP series IRF support (up to nine switches with Ethernet, two switches with storage)? Is there iSCSI support? Does the environment use front-to-back or back-to-front airflow? Does the customer prefer to have Comware OS to support their switch environment?

Virtual Connect FlexFabric adapters The HPE StoreFabric Converged Network Adapters (CNAs), shown in Figure 3-24, are dual-port adapters that provide Ethernet, iSCSI, and FC connectivity over 10GbE using both FC over Ethernet (FCoE) and Converged Enhanced Ethernet (CEE) standards.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-24 Virtual Connect FlexFabric convergence By consolidating Ethernet, iSCSI, and FC onto a converged adapter, HPE CNAs reduce the number of separate adapters and cables required for your data center and also reduce operational, power, and cooling costs while preserving existing Ethernet and FC infrastructure. HPE CNAs have also been thoroughly tested with HPE ProLiant servers and HPE ToR switches to ensure an optimal HPE FCoE solution for your data center. A FlexFabric adapter is more than a CNA because it provides standard network interface card (NIC) functions, Flex-10 NIC device capabilities, and FC or iSCSI FlexHBA capability. Important A single FlexFabric adapter cannot converge both FC and iSCSI traffic at the same time.

||||||||||||||||||||

||||||||||||||||||||

Each FlexFabric adapter contains two 10Gb/s or 20Gb/s Ethernet ports. Each port has four Flex-10 physical functions, either FlexNICs or FlexHBAs. A FlexHBA is an actual PCIe (PCI-Express, Peripheral Component Interconnect Express) physical function on the FlexFabric adapter that can be configured to handle storage traffic. The server ROM, operating system, and hypervisor recognize the PCIe function as an HBA device. You can adjust the FlexHBA transmit bandwidth in 100 Mb increments up to 20Gb.

FlexFabric physical function assignments You can assign storage traffic (FC or SCSI) as a FlexHBA only to the second physical function of each FlexFabric adapter port. HPE uses the second physical function of each port as the storage function because in a traditional CNA, it is used for storage access. The numbered physical functions (F1, F2, F3, and F4) of each port on the same FlexFabric adapter must have the same personality. This means that FC FlexHBA and iSCSI FlexHBA cannot exist on the same FlexFabric adapter at the same time. If you do not need block storage access, you can disable the FlexFabric adapter storage function and configure the second physical function as another FlexNIC function. The first, third, and fourth physical functions work only as FlexNIC devices and always have a NIC personality. You can configure the FlexFabric adapter profile through the VCM. VCM determines the personality of each of the physical functions on the FlexFabric adapter. The VCM configuration also negotiates with the Virtual Connect FlexFabric module and assigns a VLAN to each physical function. LAN traffic might also have VLAN tags from the operating system. This depends on the use of tunneling and shared uplink sets. The operating system sees each PCI physical function on the FlexFabric adapter as a conventional hardware NIC, FC HBA, or iSCSI HBA device. Each physical function advertises its VLAN assignment designated by the VCM server profile. The advertised device type and VLAN assignment steer individual traffic classes to the appropriate physical function (F1, F2, F3, or F4) on the FlexFabric adapter.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check What Protocol does not support a flat SAN?

||||||||||||||||||||

||||||||||||||||||||

SAN fabric topologies A SAN fabric topology defines the arrangement of FC switches in a fabric. There are three approaches to designing a SAN. You can implement: An HPE-standard SAN fabric topology design A subset or variation of an HPE-standard SAN fabric topology design A custom SAN fabric topology design

HPE SAN support The SAN Design Reference Guide provides the benefit of HPE engineering when building a scalable, highly available enterprise storage network. It is available at http://h20565.www2.hpe.com/portal/site/hpsc/public/psi/manualsResults/? cc=us&lang=enus&sp4ts.oid=406734&ac.admitted=1446469957277.125225703.1851288163 Regardless of which approach you use, the SAN design must adhere to the SAN design rules. Always consult the HPE SAN Design Reference Guide Part 2: Fabric Infrastructure Rules. The SAN Design Reference Guide Part 2 covers Fabric Infrastructure rules; this is broken down into different section for: FlexFabric switches Brocade B-series switches, Cisco C-series switches Including SAN fabric connectivity and interoperability.

Levels of availability Figure 3-25 characterizes data availability and indicates the supported topologies for each level.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-25 Availability levels

Single-switch fabric A single-switch fabric, shown in Figure 3-26, consists of an FC switch, server, and storage system topology; this forms the basis for all HPE-standard topologies.

||||||||||||||||||||

||||||||||||||||||||

Figure 3-26 Single-switch fabric

Meshed-switch fabric All FC switches are supported for use in a meshed fabric topology. Meshed fabric topologies, shown in Figure 3-27, typically use SAN, Fabric, or Edge switches, which support smaller incremental growth. To meet higher port-count requirements, use Core or Director switches.

Figure 3-27 Meshed-switch fabric Level 3 fabric designs provide multiple server and storage system paths to the fabric to increase availability.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Core-edge switch fabric A core-edge fabric has one or more FC switches (called core switches) that connect to edge switches in the fabric, as shown in Figure 3-28. The core switches provide high bandwidth and redundant connectivity to the edge switches. The edge switches provide user ports for servers and storage. You can also connect centralized storage (disk or tape) to the core switches if centralized access is required.

Figure 3-28 Core-edge switch fabric

Multiple fabrics and device paths This level ensures the highest availability with no-single-point-of-failure (NSPOF) protection. HPE recommends Level 4 SAN fabric designs. All FC switches are supported for use in a core-edge fabric topology. Core-edge topologies typically use SAN, Fabric, or Edge switches on the edge; and Core and Director switches in the core. When using switches with different FC maximum speed

||||||||||||||||||||

||||||||||||||||||||

capabilities (such as 1Gb/s, 2Gb/s, 4Gb/s, 8Gb/s, 16Gb/s, or 32Gb/s), HPE recommends using the higher-speed switches in the core.

Routed SAN fabrics HPE-standard fabric topologies support FC routing. FC routing enables connectivity between devices in multiple fabrics, Virtual Fabrics, or multiple VSANs. B-series (Brocade) Meta SAN A Meta SAN contains multiple B-series fabrics connected together using the B-series switches with FC routing including StoreFabric SN4000B SAN Extension Switch, 1606 Extension SAN Switches, DC Director Switch MP Extension Blades, or MP Router Blade. FC routing implements the FC routing service, which allows selective access between devices in different fabrics without having to merge fabrics. This provides a high level of isolation between fabrics. This isolation can be viewed as individual FC subnetworks within the Meta SAN. LSANs provide access to devices in different fabrics. You create LSAN zones just as you create standard zones in a single fabric. The difference is that LSAN zone definitions span multiple fabrics and therefore must be replicated on all fabrics that comprise the LSAN. C-series (Cisco) VSANs with Inter-VSAN Routing VSANs are groups of switch ports from one or more C-series switches. Each VSAN has a unique set of fabric services. Different fabric settings can be applied to each VSAN. This provides a high level of isolation between VSANs. This isolation can be viewed as individual FC subnetworks within a C-series fabric. The IVR (Inter-VSAN Routing) feature, shown in Figure 3-29, enables you to configure devices in one VSAN for access to devices in another VSAN. All C-series switches include the VSAN feature. IVR is an optional licensed software feature. There is no need for additional hardware.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-29 IFR and IVR features Brocade LSAN An LSAN is similar to an FC zone but can extend through a router to include devices in other fabrics. This configuration, which includes the physical fabrics (subnetworks), LSANs, and router, is called a Meta SAN. A Meta SAN consolidates multiple fabrics into a single entity. Routing between fabrics, called Inter-Fabric Routing or IFR, is shown in Figure 3-29. Cisco VSAN A single fabric is partitioned into several subnetworks or logical groups of switches or switch ports; this is called a VSAN. VSAN has a unique set of fabric services with independent fabric management. VSANs can share devices by using the license-enabled IVR function. ISL Trunking ISL Trunking combines multiple links between switches to form a single, logical ISL

||||||||||||||||||||

||||||||||||||||||||

with a total bandwidth of 256Gb/s. This feature enables dynamic load balancing of data across ISLs. FCIP Trunking FCIP Trunking allows multiple IP source and destination address pairs (defined as FCIP circuits) via multiples of the 1GbE, 10GbE, or 40GbE interfaces to provide a highbandwidth FCIP tunnel and failover resiliency. Each FCIP circuit supports four QoS classes as a TCP connection (Class-F, Hi, Medium, and Low priority).

Topology data access performance by SAN fabric topology HPE recommends that you implement a fully redundant fabric configuration. SAN topologies are compared in Figure 3-30.

Figure 3-30 Comparing SAN topologies To choose a SAN fabric topology, you must determine which data access type is appropriate for the environment. The data access types include: Local (one to one)—Data access between a local server and a storage system

Technet24

||||||||||||||||||||

||||||||||||||||||||

connected to the same switch Centralized (many to one)—Data access between multiple, dispersed servers and one centrally located storage system Distributed (many to many)—Data access between multiple, dispersed servers and storage systems

External connections If you use external storage, you can use a direct-connect technology (usually with an external SAS —Serial Attached SCSI — cable) or FC, which can be connected directly or through an FC switch. You can also use iSCSI, which travels over regular Ethernet hardware and cables. The common issue is cost. FC is lossless and performs at the maximum speed of the wire, assuming the disks can push and pull that much data. iSCSI has the standard TCP overhead. It can be reduced using jumbo frames, but it is still there. However, with 10GbE connections, the overhead is probably of little concern. Another option is SMB 3.0. Similar to iSCSI, this works over regular Ethernet. FCoE SAN Fabrics HPE-standard fabric topologies can integrate with FCoE technology. FCoE is deployed in existing Ethernet and FC environments providing convergence at the server and fabric edge using CNAs and FCoE CN switches. This allows FC to use 10GbE networks and preserve the FC protocol. Simultaneously, the 10GbE network supports Ethernet data, thus creating a single converged network for Ethernet and FC. The convergence of Ethernet and FC provides the same level of connectivity that each technology provides separately but requires: 50% fewer switches in each server rack (providing the same level of connectivity as four switches per rack with separate Ethernet and FC switches, two for each technology) 50% fewer adapters per server 75% fewer cable connections HPE FCoE fabric-edge solutions deploy FCoE technology in Ethernet and FC environments with the benefit of convergence at the server and fabric edge using CNAs and converged network switches. These solutions also provide investment protection by allowing FCoE to be integrated with existing HPE FC SANs. FCoE technology allows you to converge Ethernet and FC technology, providing

||||||||||||||||||||

||||||||||||||||||||

significant cable, adapter, and switch consolidation. All FC topologies supported by HPE are integrated with FCoE at the FC fabric edge. HPE also supports end-to-end FCoE solutions. HPE-standard fabric topologies can integrate with FCoE technology. With HPEstandard SAN fabric topologies, you can: Create a SAN fabric for each department or application in the organization. Perform centralized management and backups. Update a SAN fabric to accommodate changing capacity or data access needs and convert to another SAN fabric topology as needed. Connect devices over long distances using extended FC or IP connections. Connect multiple SAN fabrics using routing technology.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check Name the three approaches to an HPE SAN Design.

||||||||||||||||||||

Activity: Using the SAN Design Reference Guide Instructions: Use the “SAN Design Reference Guide” at: https://support.hpe.com/hpsc/doc/public/display?docId=c00403562 for this activity. Scenario: Your customer requires an FC SAN with high availability to support 20 hosts, StoreOnce, 3PAR StoreServ 8200. Answer the following questions: –

Fabric design data availability: What fabric design would you suggest and why?

–

Calculating for data availability levels: What is the hardware cost and the number of available ports required?

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Virtualization management with HPE OneView HPE OneView is a scalable, resource-oriented solution focused on the entire life cycle —from initial configuration to on-going monitoring and maintenance—of both physical and logical resources (resources shown in Figure 3-31): Physical resources are objects you can touch, such as server hardware, interconnects, ToR switches, enclosures, storage systems, and racks. Logical resources are virtual objects such as templates or groups that when applied to physical resources, provide a common structure across your data center.

Figure 3-31 Data center components HPE OneView provides several software-defined resources, such as server profile templates (SPTs), to enable you to capture the best practices of your experts across a variety of disciplines, including networking, storage, and hardware configuration. HPE OneView can be deployed as a standalone converged management solution or incorporated across a wider environment. This management solution is a strategic component in the HPE strategy for converged infrastructure, cloud, and an open SDDC environment. An SDDC enables customers to manage their infrastructure as pools of resources, streamline and automate administrative and operational tasks, achieve significant staff efficiencies and productivity increases, roll out new technologies and applications faster, and lower the costs associated with downtime and regulatory risks. HPE OneView features:

Technet24

||||||||||||||||||||

||||||||||||||||||||

A single, integrated management environment for a converged infrastructure that enables IT teams to work and collaborate in a more natural and automated way An intuitive interface and powerful search capabilities Unified Representational State Transfer (REST) API and a Java software development kit (SDK) for integration with enterprise management tools and applications In addition to remote management and environmental management of the data center, HPE OneView enables efficient, reliable, nondisruptive, and simple firmware management across the data center with firmware baseline capabilities.

||||||||||||||||||||

HPE OneView Features of HPE OneView, shown in Figure 3-32, include: Profile-driven automated volume provisioning with StoreVirtual. –

HPE StoreVirtual VSA software-defined array.

–

HPE StoreVirtual 3200 series hardware appliance.

||||||||||||||||||||

iSCSI boot and data volume access configuration. Scheduled firmware and OS drivers’ update. External firmware repository—provides unlimited storage space for SPPs. Supports multiappliance environments with single external appliance. Extending software-defined intelligence across the server portfolio. OneView Advanced Support for Apollo 2K, 4K, 6K and ML350. OneView Standard Support for Superdome X and remaining ML servers (beyond ML 350).

Figure 3-32 Features of HPE OneView Now includes advanced management of HPE ProLiant ML350 Servers and Apollo 2000/4000/6000; also includes standard monitoring and reporting of HPE Superdome X and the rest of the HPE ProLiant ML Servers, as well as Red Hat KVM hypervisor support. Virtual Connect 16Gb FC offers bidirectional port mirroring, throughput utilization statistics, detailed connector data, and digital diagnostics. Role-based access provides logical grouping of systems so that IT staff can focus on the systems they are responsible for managing; includes storage, network, and converged administrative roles; adds the role of firmware operator to provide extra security.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE OneView Global Dashboard unifies monitoring of HPE platforms at scale; aggregates contents from up to 20 HPE OneView appliances; and manages up to 12,800 devices.

Consumer-inspired user experience From the dashboard, administrators can assess the entire data center in seconds. Whether the customer has 20 devices or 20,000 devices, the view is equally simple. Major functions of HPE OneView and mapping to most common data center operations are shown in Figure 3-33.

Figure 3-33 Consumer-inspired user experience of HPE OneView To help customers find key information in milliseconds instead of hunting through offline records, HPE created Smart Search. Smart Search is built into almost every task, so administrators have immediate access to the event, device, or task information they need. Map View combines device status with relationships, so customers can see how things are connected from the data center to the device. This helps customers quickly find, triage, and fix problems in seconds or minutes. With templates, IT teams can work together to create profiles of configurations and procedures. Any configuration or process can be designed by the experts and captured in software for later use by everyone. Activity Feed updates users with alerts, jobs, and projects in the data center. With this advanced functionality, communication becomes more connected and happens in real time.

||||||||||||||||||||

||||||||||||||||||||

HPE OneView licensing HPE OneView supports the following license types: HPE OneView Standard, A free license that enables monitoring of supported HPE hardware. HPE OneView Advanced, A license that enables all HPE OneView management features for supported HPE hardware. There are two types of HPE OneView Advanced licenses: HPE OneView Advanced w/o integrated Lights Out (iLO), for server hardware that has an existing iLO Advanced license, or that does not require the features enabled by iLO Advanced. HPE OneView Advanced with iLO Advanced, includes an iLO Advanced license for the server hardware, which enables advanced management features.

Smart Update Tools features Smart Update Tools (SUT) is an operating system utility for HPE OneView that enables an administrator to perform online firmware and driver updates. SUT polls HPE OneView every five minutes for new requests, processes those requests, and provides HPE OneView with a status. HPE OneView posts the progress in the Firmware section of the Server Profile page. SUT installs updates in the correct order and ensures that all dependencies are met before starting an update. If there are unmet dependencies, SUT prevents the installation and notifies the HPE OneView administrator that the installation cannot continue because of a dependency.

HPE OneView for storage management With a variety of internal and external storage options, the HPE Synergy Architecture can provide fluid pools of resources for any workload, from traditional applications to cloud-native applications. Synergy Storage Modules can directly be composed as DAS supporting everything from traditional applications such as Exchange or database applications to Hadoop Analytics. Using StoreVirtual VSA to create virtualized clusters supported by Synergy Storage Modules allows customers to take advantage of tiering for performance on a highly dense, scalable storage platform with the flexibility required for their VMs and VDI solutions.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ, as an external-to-the-frame option, but still a part of the overall Composable Storage family continues to deliver the same value proposition that many customers have found so compelling for accelerated and assured service levels for virtualization and IT as a Service, or more traditional use cases around applications such as SAP or Oracle. OneView orchestrated storage provisioning Accelerates provisioning for 3PAR, StoreVirtual, Synergy D3940. Storage defined in Profiles or Profile Templates for Control AND Agility. Storage provisioned when profile assigned to the infrastructure/workload. OneView automated storage access Automated configuration—SAS fabric or FC/FCoE Zoning. Nondisruptive storage reconfiguration. Composable storage can follow profile migration across servers.

The best of control and agility with storage templates HPE OneView implements the Composable Infrastructure vision, as shown in Figure 334, and provides the ability to automate volume provisioning across StoreServ and StoreVirtual in a consistent fashion, giving users the flexibility to create storage pools and use templates for storage provisioning. Other features include: Create specific templates that set up service levels and policies for each environment based on needs. Deploy templates into the infrastructure. Templates provide configuration consistency, reliability and control.

||||||||||||||||||||

Figure 3-34 Three pillars of Composable Infrastructure

StoreVirtual automation—HPE OneView HPE OneView: Fully Automates Server + Storage Provisioning for iSCSI, FC, SAS, as shown in Figure 3-35. –

Support for StoreVirtual-specific resources, including Storage systems, Storage pools, Volume templates, and Volumes.

Policy-based deployment eliminates errors. –

SPT support for StoreVirtual/iSCSI-managed storage.

–

iSCSI boot and data volume creation and access configuration.

–

Simplified boot from iSCSI-managed volume.

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 3-35 Storage automation in HPE OneView For a user, the OneView storage templates will be used to abstract the storage hardware so that the user can specify a class of service. This applies to both the HPE StoreVirtual hardware appliance and HPE StoreVirtual software-defined array (VSA). For StoreVirtual, a volume is attached to a server using the iSCSI protocol for both boot volumes and data volumes.

Resource model diagram The UI and REST APIs are organized by resource. The documentation for the UI and REST APIs is also organized by resource. The complete list of resources is included in the HPE OneView REST API Reference in the online help. HPE OneView uses a resource model, shown in Figure 3-36, that reduces complexity and simplifies the management of your data center. This model provides logical resources, including templates, groups, and sets, that when applied to physical resources provides a common structure across your data center.

||||||||||||||||||||

||||||||||||||||||||

Figure 3-36 Storage automation in HPE OneView The UI distinguishes between physical and virtual resources by using certain actions. For example: You can create, delete, or copy a logical resource, but not a physical resource. You can add or remove a physical resource. Note For more information on resources: http://h20566.www2.hpe.com/hpsc/doc/public/display? sp4ts.oid=5410258&docLocale=en_US&docId=emr_na-c05098313 Several REST API resources are related to the appliance and appliance settings.

Management platform built on the unified REST API HPE OneView features an open development platform, shown in Figure 3-37, designed to adapt rapidly to business needs. This programmable platform, built on the unified

Technet24

||||||||||||||||||||

||||||||||||||||||||

REST API, allows enterprises to scale beyond data center walls to the cloud.

Figure 3-37 Programmable interface of HPE OneView REST uses four basic commands (create, read, update, and delete) that enable administrators to define and radically simplify the most complex processes. Simply put, a REST call is an HTTP request to a server. REST has become a popular communication protocol on service-oriented architecture styles. Applications conforming to REST constraints are known as RESTful. The HPE unified REST API can be used to automate, integrate, and customize anything that can be done from the user interface using a favorite scripting or programming language. The unified REST API provides a programmable interface and lightweight data model specification that is simple, remote, secure, and extensible. It is designed for those who want to access the full power of the management architecture. APIs provide a

||||||||||||||||||||

||||||||||||||||||||

simpler, stateless, and scalable approach, so users can easily integrate, automate, and customize on their own. The storage system API enables administrators to bring a storage disk array under HPE OneView management.

HPE OneView: Your infrastructure automation engine HPE OneView is your infrastructure automation engine to simplify operations, increasing the speed of IT delivery for new applications and services. Through software-defined intelligence, HPE OneView brings a new level of automation to infrastructure management by taking a template-driven approach to provisioning, updating, and integrating compute, storage, and networking infrastructure. Designed with a modern, standard-based API and supported by a large and growing partner ecosystem, HPE OneView also makes it easy to integrate powerful infrastructure automation into existing IT tools and processes. Deploy infrastructure faster: Template-based automation enables IT generalists to rapidly and reliably provision resources in response to application owner requirements. Simplify operations: Agentless monitoring, online firmware updates, and a new Global Dashboard deliver streamlined lifecycle operations at scale. Increase productivity: The unified API allows developers and ISVs (independent software vendors) to harness the power of HPE OneView unified infrastructure automation with application and IT service delivery.

HPE OneView licensing Both licensing modes can be used on the same appliance: HPE OneView Advanced Standalone license with iLO Advanced Standalone license w/o iLO Advanced License sold with new ProLiant DL server with iLO Advanced License sold with new ProLiant DL server w/o iLO Advanced Software for new BladeSystem c7000 with 16 licenses with iLO Advanced Software for new ProLiant BL blade server with iLO Advanced Upgrade from HPE OneView from HPE VCEM, HPE Insight Control, or HPE iLO

Technet24

||||||||||||||||||||

||||||||||||||||||||

Advanced w/o iLO Advanced HPE OneView Advanced includes 3-year Technical Support and Software Updates. HPE OneView Advanced licenses (except upgrade) include right-to-use HPE Insight Control. HPE OneView Standard HPE OneView Standard can be used by downloading the HPE OneView management appliance. Annual 9x5 support is available at additional charge.

Managing storage resources with HPE OneView HPE OneView provides automated, policy-driven provisioning of storage resources. With HPE OneView, you can view and manage your storage system and storage pools. You can add existing volumes and create new volumes, and you can create volume templates to provision multiple volumes with the same configuration. Figure 3-38 shows the HPE OneView Dashboard with several storage resources.

Figure 3-38 Dashboard of HPE OneView

Storage systems A storage system (or storage array) is hardware that contains multiple storage disks. Bringing SAN storage systems under management of the HPE OneView appliance

||||||||||||||||||||

||||||||||||||||||||

enables you to add and create logical storage spaces, known as volumes. You can then attach volumes to server profiles through volume attachments. This enables the server hardware assigned to the server profiles to access the SAN storage system. When adding a storage system, you must choose a domain on the storage system. You can then select storage pools from that domain on the storage system to add to the appliance. After you add storage pools, you can assign FC networks to the storage ports associated with the storage system. HPE OneView monitors the health status of storage systems and issues alerts when there is a change in status. It also monitors the connectivity status of storage systems. If HPE OneView loses connectivity with a storage system, an alert is displayed until connectivity is restored. HPE OneView attempts to resolve connectivity issues and clear the alert. If it cannot, you must resolve the issues and use the storage systems screen to manually refresh the storage system and synchronize it with the HPE OneView appliance. HPE OneView also monitors storage systems to ensure that they are synchronized with changes to hardware and configuration settings. However, changes to storage systems made outside the appliance (such as changing credentials) might cause the storage system to lose synchronization with the appliance, in which case you must manually refresh the storage system.

HPE OneView and HPE 3PAR StoreServ systems HPE OneView is fully integrated with server profiles so that you can manage new or existing HPE 3PAR StoreServ infrastructure. Server profiles are one of the features of HPE OneView that enable you to provision converged infrastructure hardware quickly and consistently according to best practices. A server profile captures key aspects of a server configuration in one place, including firmware levels, Basic Input/Output System (BIOS) settings, network connectivity, boot order configuration, iLO settings, and unique IDs. When you import existing HPE 3PAR StoreServ storage systems and storage pools, HPE OneView can quickly create volumes. You attach volumes to HPE OneView server profiles to make them available to server hardware. You can attach private or shared volumes to HPE OneView server profiles to enable automated boot target configuration and to move direct-attach profiles across enclosures. HPE 3PAR StoreServ volume attachment is an automated method to provision host

Technet24

||||||||||||||||||||

||||||||||||||||||||

access to volumes on a storage system and to automate FC zoning. The server administrator can attach one or more volumes to a server profile. When the server profile is assigned to a server, HPE OneView orchestrates the host creation and VVol export on the 3PAR StoreServ system and configures any required FC zones for fabric-attach networks. Switched fabric, direct-attach, and VSAN SAN topologies are supported for communication between devices: Brocade Network Advisor (BNA) manages Brocade FC SAN configurations (for example, SANs, LSANs, and Meta-SANs). HPE 3PAR StoreServ systems connected directly to an enclosure via FC are discovered dynamically by HPE OneView. You can also add SAN managers to bring their managed SANs under management of HPE OneView. Managed SANs can be associated with FC networks on the appliance to enable automated zoning and automatic detection of FC connectivity. When a SAN manager has been added, HPE OneView can assist with assigning the storage system host ports to FC networks automatically. The identified ports and their associated FC networks can be checked from the HPE OneView storage system menu.

Volumes Volumes are logical storage spaces that are provisioned in storage pools on storage systems. Volumes are associated with server profiles through volume attachments. You can add a storage volume that exists on a storage system under management of HPE OneView. Attaching a server profile to a volume gives the server hardware to which the server profile is assigned access to storage space on a storage system. Volumes can be created with either the server or storage role. Using volume templates, you can create multiple volumes with the same configuration. Storage volume templates enable you to enforce certain volume parameters when the storage volumes are being created. Storage volume templates are not required in order to provision storage volumes. You also can create a storage volume without a storage volume template, but you must specify the storage volume provisioning attributes. Before you create a storage volume, you select a storage system and associated storage pool. After you create a storage volume, you can add it to a server profile. HPE OneView can import existing, unexported volumes for assignment to server profiles. The internal 3PAR StoreServ volume ID will be required. You can add existing volumes either in the storage systems or volumes view. HPE OneView also

||||||||||||||||||||

||||||||||||||||||||

enables you to update and delete a storage volume.

Volume provisioning directly from HPE OneView HPE OneView supports the automated volume provisioning and Brocade fabric zoning of HPE 3PAR StoreServ systems. To provide additional automated HPE 3PAR StoreServ automation, BNA must be installed and available to be configured as a SAN manager. SAN managers enable you to use HPE OneView to automate FC switch configuration and to assist with validating the expected storage connectivity model when importing a storage system. Note A version of BNA SAN Professional is available for SMBs at no charge. However, this version lacks SMI-S/WBEM, which HPE OneView uses for auto-zoning. Because of this limitation, the no-charge version of Network Advisor SAN Professional is not supported by HPE OneView. Adding a supported SAN helps align managed SAN fabrics, FC networks defined by HPE OneView, and HPE 3PAR StoreServ front-end ports. This provides a simplified way to guarantee that connected storage system ports are available on the expected HPE OneView–managed FC networks that will be assigned to server profiles. Note If HPE OneView cannot connect to a supported HPE 3PAR StoreServ array, verify that the Web Services Application Program Interface (WSAPI) service (showwsapi) is started on the HPE 3PAR StoreServ array. There is a known condition where the WSAPI service needs to be restarted with a stopwsapi –f followed by startwsapi.

Storage pools A storage pool is an aggregation of physical storage disks in a storage system. Storage systems contain information about the storage ports through which they can be accessed. You can provision volumes from storage pools. You can choose one or more storage pools when adding a storage system to HPE OneView. Storage pools are created on a storage system using the management software

Technet24

||||||||||||||||||||

||||||||||||||||||||

for that system. You cannot create or delete storage pools from the appliance—you can only add or remove them from management. After you add storage pools, you can provision volumes on them.

HPE OneView domains An HPE OneView domain is a Converged Infrastructure management framework concept. You cannot create additional HPE OneView domains, but the appliance itself is a single domain construct. An HPE OneView domain consists of one or more logical interconnect groups, uplink sets, and networks. This domain helps to define how server profiles consume these resources. When defining a network, the domain will be available for consumption by either a logical interconnect group, logical uplink set, logical interconnect (for one-off configuration requirements), or server profiles (for internal-only networks). Note To access HPE OneView QuickSpecs, go to https://www.hpe.com/h20195/v2/GetPDF.aspx/c04111367.pdf

Deploying applications using rapid automation and growth After the initial configuration is complete, you can access HPE OneView for VMware vCenter. HPE OneView for VMware vCenter enables you to: Create and extend datastores. Deploy or clone VMs. Instantiate VMs from templates. Monitor capacity utilization, space savings, and more within one platform. Advanced features are accessible via the StoreVirtual Centralized Management Console. You can: Access application-integrated snapshots. Set up multiple sites in vMSC environments. Perform replication to other StoreVirtual systems.

||||||||||||||||||||

||||||||||||||||||||

What’s new Support for HPE ProLiant and HPE Synergy Gen10 Servers. Support for HPE OneView. Support for HPE Synergy Image Streamer for patching ESXi cluster images. Similar to the initial deployment, HPE OneView Instant On guides users through the process of adding another appliance to (or “expanding”) an installation. When this happens, the storage configuration matches the compute cluster. HPE OneView Instant On expands the installation by performing a system health check, defining configuration of new appliances, and rolling out configuration and adding resources to clusters.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check What are the three main components of the HPE OneView domain?

||||||||||||||||||||

||||||||||||||||||||

Activity: HPE OneView Licensing Guide video Instructions: Watch the “HPE OneView Licensing Guide video” at: https://www.hpe.com/h22228/video-gallery/us/en/sss/solutions/convergedinfrastructure/7B411836-35A4-462D-982E0DD9447C2BF6/hpeoneviewlicensingguidevideo/video/. Notes:

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE Converged Systems HPE ConvergedSystem is a portfolio of products that combine servers, storage, networking, software, and services for specific workloads. ConvergedSystem solutions use a common converged infrastructure architecture, the same foundation used for all HPE servers, storage, and networking products. These solutions are easy to procure and support, and all parts are managed from one system. These IT products, shown in Figure 3-39, are designed to lower the cost and complexity of data center operations and maintenance by pulling the IT components together into a single resource pool. This makes them easier to manage and reduces the time required to deploy applications.

Figure 3-39 HPE Converged Systems HPE Converged Infrastructure solutions pool resources so that they can be shared across different applications and managed using a standardized and secure platform. The convergence of server, storage, and networking can help organizations reduce spending on equipment maintenance and management. HPE Converged Architecture 700 offerings are modular, repeatable, and scalable building blocks designed for workloads. It is infrastructure delivered as one integrated stack by HPE Certified Channel Partners, and with verified reference architectures thoroughly tested by HPE engineering, to save your time and resources. HPE ConvergedSystem 500 and 900 for SAP HANA These scalable, performance-optimized systems reduce complexity and risk while providing the fastest path to value for SAP HANA. Features include built-in high

||||||||||||||||||||

||||||||||||||||||||

availability and broad scalability to grow as your data grows. HPE ConvergedSystem 300 for Microsoft® Analytics Platform Designed to deliver high performance, it is a purpose-built appliance that integrates in-memory performance, SQL Server Parallel Data Warehouse (PDW), and hosted desktop infrastructure (HDI) seamlessly into one system. Backed by end-to-end integrations and deployment services, it provides customers with design flexibility for large amounts of data, up to 6 PB. Note For more information about ConvergedSystem options, open the hyperlink: https://www.hpe.com/us/en/integrated-systems/converged.html Many IT organizations have launched server virtualization and cloud computing projects to achieve improved service delivery and increased IT agility. However, for some organizations, ambitious virtualization, cloud, and infrastructure initiatives have led to lengthy projects, increased management complexity, and higher operational costs. Hyperconvergence provides many of the benefits of a virtualized data center in a compact, cost-effective system that is easy to deploy, manage, and support. With hyperconvergence, everything is preconfigured, preinstalled, ready to run in minutes, and designed to scale—with no need for a separate SAN. With turnkey hyperconverged systems, anyone from IT generalists to experienced admins can quickly deploy complete IT environments, going from power-on to provisioning in as little as 15 minutes. And they can do it all without the assistance of outside system integrators. Hyperconverged systems scale out in a linear manner. Just add another system and seamlessly add new nodes to a common cluster. The hyperconverged architecture integrates grid software that automatically discovers and adds new nodes to the cluster, delivering additional compute and storage resources with the addition of each new module. System management is straightforward. There is no need for a team of specialists with expertise in virtualization, storage, servers, and networking. Compared to systems that require a lot of discrete components, including an associated SAN with its own switches, the simplified nature of hyperconverged systems can help you reduce capital and operational expenses. Everything is in one easily managed and serviceable, smallfootprint box that helps pack more computing and storage capacity into customers’ valuable data center space. A single vendor can support the entire system—including hardware and software.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Hyperconverged infrastructure natively integrates compute and storage into a single x86-based server that can be deployed in scale-out clusters. It reduces power and space and eliminates storage complexity. Built-in virtualization makes infrastructure truly invisible, shifting focus back to workloads. Hyperconverged systems enable hybrid infrastructure that is poised to respond faster and more effectively to business needs.

HPE ConvergedSystem 700 The HPE ConvergedSystem 700, shown in Figure 3-40, is purpose built, factory integrated, and optimized to meet all of your virtualization, hybrid cloud, and workplace productivity needs with faster time to value. As an open, modular, and scalable converged system, the HPE ConvergedSystem 700 is simple to deploy, manage, and support. Open, flexible, and scalable, ConvergedSystem 700 ships with factoryintegrated server, storage, networking, and management components—all preconfigured to deliver speed and simplicity. The system also includes integrated lifecycle management compliance, all the way to the workload level, as well as solution-level support with single-vendor accountability. The new HPE FlexFabric 5940 is optimized for higher-performance server connectivity, convergence of Ethernet and storage traffic, the capability to handle virtual environments, and low latency. New option for the Cisco Nexus 9000 Series Switch delivers proven high performance and density, low latency, and exceptional power efficiency in a broad range of compact form factors. Support for HPE OneView—The infrastructure automation engine on HPE ConvergedSystem 700 that helps simplify operations, increasing the speed of IT delivery for new applications and services. HPE Helion CloudSystem 10 support—Gain more flexibility, more visibility, and more freedom with the latest release of HPE Helion CloudSystem. HPE 3PAR All-inclusive Software Licensing. Every 3PAR system now ships with every feature you need to manage a single array. For multiple systems, add a single Multisystem Software option.

||||||||||||||||||||

||||||||||||||||||||

Figure 3-40 HPE ConvergedSystem 700

Hyperconvergence—Support for a hybrid infrastructure Hyperconverged systems enable you to: Rapidly develop and deliver new services and growth. Create new revenue streams and tap into new markets. Eliminate infrastructure and management complexity. Increase overall efficiencies and lower operational costs. This approach enables a high-velocity business model that delivers faster time to value, reduces both costs and risks, and positions your organization to move toward cloudbased models that increase your competitive advantage. The HPE Hyper Converged system is much more than a combination of compute, storage, and networking components, as shown in Figure 3-41. It incorporates automated management to radically simplify everyday tasks, reduce OPEX, and improve operational agility.

Technet24

||||||||||||||||||||

Figure 3-41 Hyperconvergence benefits Benefits that hyperconverged systems can provide include: Business benefits—From a business perspective, the benefits of hyperconverged systems include reduced upfront capital and operational costs, thanks to fewer components than are required in conventional systems; a small footprint; simplified management; and efficient power and cooling. Customers do not need specialists to maintain and operate a hyperconverged system. An IT generalist will most often be able to bring it online and provide daily management. These systems provide: –

Faster deployment

–

Linear scalability

–

Straightforward management

||||||||||||||||||||

IT benefits—IT professionals do not have to manage disparate server, storage, and networking systems. The infrastructure includes built-in resiliency and is easy to set up, learn, service, and scale. Four powerful servers, all-inclusive enterprise-class data services, and inherent data protection provide the same level of service as traditional IT solutions.

||||||||||||||||||||

||||||||||||||||||||

Overall benefits—Customers on both the business and IT sides can benefit from the increased agility of IT systems that can be brought online easily and scaled out in a linear manner. When new resources are needed, a hyperconverged system can be up and running in minutes.

What is HPE Hyper Converged? Hyperconverged models are the fastest way to an efficient virtualized environment. There is no assembly needed, and all components are pre-integrated, as shown in Figure 3-42. Examples of hyperconverged systems include:

Figure 3-42 HPE Hyper Converged system HPE Hyper Converged 250—Pre-configured and integrated virtualized blocks of infrastructure can be ready in 15 minutes, scale out linearly, and easily integrate with a centralized data center. The HPE HC 250 is a compact 2U/4-node virtualized platform of powerful compute and resilient storage managed as one from a single interface and optimized to handle a variety of workloads—from on-demand IT infrastructure to VDI. HPE Hyper Converged 380 combines HPE ProLiant DL 380 Servers, VMware® virtualization software, and HPE management tools to deliver a system that is quick to install, easy to manage, and expand. If needed, it quickly increases your system from 2 nodes to 16 by adding more nodes. Management software automatically recognizes and integrates new nodes into the cluster. Select one of three workload configurations, and then customize the storage, server, and networking to deliver an optimal (or ideal) configuration for your business needs. Perform VM and live hardware firmware management from the integrated HPE OneView user experience. Hyper Converged 250 for Microsoft CPS Standard offers instant compute and shared storage cluster serving both applications and data services managed within the

Technet24

||||||||||||||||||||

||||||||||||||||||||

Microsoft System Center console. It also delivers a self-service cloud environment with services and automation that helps IT streamline deployment of Windows and Linux workloads. This hyperconverged and Microsoft Azure®-consistent cloud in a box lets you deploy an infrastructure that enables you to reduce cost and deliver IT services faster. HPE SimpliVity 380 is designed for IT leaders struggling to achieve the agility and economics of the cloud, with the control and governance of on-premises IT. HPE SimpliVity 380, available on HPE ProLiant DL380 Gen9 Servers, is a compact, scalable 2U rack-mounted building block that delivers server, storage, and storage networking services. The HPE SimpliVity 380 delivers an all-flash solution delivering predictable and near-linear performance at near-wire speeds while reducing latency.

The evolution of hyperconvergence By consolidating the entire stack, HPE SimpliVity is able to provide all IT and data services for virtualized workloads on a powerful HPE ProLiant server. And this is all enabled by our software, the HPE SimpliVity Data Virtualization Platform. The evolution toward SimpliVity is shown in Figure 3-43.

Figure 3-43 The evolution of hyperconvergence

||||||||||||||||||||

||||||||||||||||||||

HPE SimpliVity For IT leaders struggling to achieve the agility and economics of the cloud, with the control and governance of on-premises IT, HPE SimpliVity 380, shown in Figure 3-44, delivers a powerhouse hyperconverged solution running some of the world’s most efficient and resilient data centers.

Figure 3-44 HPE SimpliVity DL380 HPE SimpliVity 380, available on HPE ProLiant DL380 Gen10 Servers, is a compact, scalable 2U rack-mounted building block that delivers server, storage, and storage networking services. It also delivers a complete set of advanced functionalities that enables dramatic improvements to the efficiency, management, protection, and performance of virtualized workloads. Features include: Factory integrated hyperconverged solution directly from HPE. Featuring HPE OmniStack software running on HPE DL380 Servers. Intel® Xeon® E5-2600 v4 Processor (Broadwell) support All-Flash Storage. What HPE SimpliVity does is take the entire legacy stack on the left of Figure 3-45: everything including servers, storage switch, HA-shared storage (High availability), and SSD arrays, backup, WAN optimization, and data protection apps—and we consolidate it all into a 2U appliance.

Technet24

||||||||||||||||||||

Figure 3-45 HPE SimpliVity components HPE SimpliVity HyperGuarantee Features of the HPE SimpliVity HyperGuarantee include: HyperEfficient: save 90% capacity across storage and backup combined HyperProtected: under one minute to complete a local backup or local restore of a 1TB VM HyperSimple: Three clicks to back up, restore, move, or clone a VM from a single console HyperManageable: under one minute to create or update backup policies for 1000s of VMs across dozens of sites HyperAvailable: add or replace HPE SimpliVity systems with zero downtime for local or remote sites –

Zero disruption to local or remote SimpliVity backups

–

Zero reconfiguration of SimpliVity backup policies for local or remote sites

–

Zero re-entry of IP addresses in remote sites

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Note HPE SimpliVity HyperGuarantee applies to current HPE SimpliVity and HPE OmniStack products and the new HPE SimpliVity 380 product.

Hyperconverged use cases Remote locations Businesses with remote locations benefit from standardized systems that are easy to deploy, manage, and scale, with the work handled by IT generalists. To keep pace with growing and sometimes unpredictable workloads, remote sites need systems that can scale quickly and seamlessly. And to maintain business continuity, they need systems that provide centralized backup, recovery, and replication of data, following standard configurations and corporate processes. Hyperconverged systems meet all of these needs. For example, if a company has offices located around the world, a common solution is ideal. It allows the company to standardize processes for deploying, managing, and supporting its systems in remote locations. The company can also standardize its backup and recovery processes and leverage the existing resources in the corporate data center. Developers in different locations also benefit from the capabilities of the hyperconverged system, since each development team can have its own dedicated VM for projects. And as their projects grow, developers can quickly add infrastructure resources to maintain excellent response times. The simplicity of hyperconverged systems allows IT generalists to handle the day-to-day systems maintenance and management work in the remote locations. That helps the company reduce operating costs and meet the full range of computing and storage needs at the remote sites. Lines of business In many large organizations, lines of business or internal departments need a simple way to handle application services without a lot of IT overhead and without reducing performance expectations. For example, your customer’s finance department might routinely deal with quarter-end report generation. As the data grows, the reports take longer and longer to process. This finance department also has new projects in the planning stage, including an advanced accounting system that may need to be rolled out within the quarter. A hyperconverged system is a perfect solution to support new applications or services because it can be easily ordered and set up quickly with pre-scripted configuration tools. The ability to quickly roll out a fully functioning IT environment could allow the company to deploy Technet24

||||||||||||||||||||

||||||||||||||||||||

its new accounting system before the end of the quarter, generate reports in a timely manner, and avoid the end-of quarter chaos. Midsize businesses Hyperconverged systems can also support midsize businesses. For example, your customer is a midsize business that needs to expand its server and storage capacity and wants to avoid taking on a lengthy, costly IT project. Hyperconverged systems are essentially turnkey data centers in a box that can be easily procured and deployed in minutes. Another example is a hospital that wants to provide its clinical staff members with access to applications and data in a secure manner, whether they are on the hospital floor or working from a remote location. The hospital can deploy hyperconverged systems to run a VDI environment. The VDI solution gives hospital staff access to applications and data that reside on a centralized system. This allows the hospital to meet the access and performance requirements of clinical staff in an efficient manner and maintain patient confidentiality. As the volume of stored data grows over time, the hospital can easily expand its VDI environment by adding additional hyperconverged systems, which seamlessly expand a common pool of resources in a virtualized, software-defined environment. The hospital also benefits from the small footprint of the systems, which makes it easier to scale out infrastructure in a space-constrained data center.

HPE Composable Infrastructure HPE Synergy is a new category of infrastructure designed to bridge traditional and cloud‐native applications through the implementation of Composable Infrastructure. Composable Infrastructure, shown in Figure 3-46, helps organizations achieve important objectives related to infrastructure, objectives that are difficult—if not impossible—to achieve via traditional means, such as the ability to do the following: Deploy quickly with simple flexing, scaling, and updating. Run workloads anywhere—on physical servers, on virtual servers, or in containers. Operate any workload on which the business depends, without worrying about infrastructure resources or compatibility. Ensure the infrastructure is able to provide the right service levels so business can stay in business.

||||||||||||||||||||

||||||||||||||||||||

Figure 3-46 HPE Composable Infrastructure So, what is a Container? The easiest way to understand containers is by comparison to VMs. VM hypervisors—Hyper-V and vSphere—are based on emulating every aspect of server hardware, making them greedy in terms of the demands on the underlying physical server. In addition, each VM hosts its own copy of an operating system (OS)— Windows or Linux. This combination of emulation plus the need to host its own local OS is called the hypervisor performance penalty and can be as high as 20% depending on the use case. Containers only virtualize the application environment, not the hardware—a much less complex task. Containers also do not “contain” a local OS: a container running on a server shares the single underlying Linux OS. That means they are much more efficient than hypervisors in terms of system resources. Also, since there is no hypervisor, your customer is probably moving to reduce spending on software license fees. Container technology is mostly open source and Docker, Inc. is a big player. While they did not invent Container technology, they did invent this particular form that is massively gaining traction in the marketplace.

Benefits of Composable Infrastructure Technet24

||||||||||||||||||||

||||||||||||||||||||

In a stateless architecture, such as that provided by HPE Synergy, state information is independent of hardware and can be easily shifted to a new location, along with any workloads that are dependent on that state. Benefits of HPE Composable Infrastructure are shown in Figure 3-47.

Figure 3-47 Benefits of HPE Composable Infrastructure Composable Infrastructure is made up of 3 main components: Fluid resource pools, pools of compute, storage, and fabric work as a single structure, ready to boot up for any workload. Effortlessly meet each application’s changing needs by allowing for the composition and re-composition of single blocks of disaggregated compute, storage, and fabric infrastructure. Software‐defined intelligence—this means that HPE Synergy puts the intelligence right into the infrastructure and uses workload templates to speed up deployment and create frictionless change eliminating unnecessary downtime. Unified API provides a single interface to discover, search, inventory, configure, provision, update, and diagnose the Composable Infrastructure. A single line of code enables full infrastructure programmability and can provision the infrastructure required for an application.

The versatility of Composable Storage via HPE OneView HPE Synergy, featuring the HPE Synergy Composer software, gives customers the

||||||||||||||||||||

||||||||||||||||||||

flexibility to compose storage for whatever it is they want to do—all with a single infrastructure. As simple as using a credit card to buy services in the cloud, composable storage is provisioned with just a few clicks but without the exasperating experience of being charged for things you did not expect. Composable storage and Synergy make your whole IT operation more efficient by significantly reducing the time that might otherwise be spent provisioning storage. Even SAN management, one of the most vexing processes in traditional infrastructures is made simple with Synergy. Your IT staff can rely on Synergy storage automation to do things quickly and correctly, so they can move on to work on more interesting projects. Whether you use high-density HDDs and SSDs internal to Synergy Frames or external 3PAR storage, Synergy composed storage can be used by any of applications running on VMs and bare metal servers anywhere in your Synergy infrastructure. That reduces the amount of time you need to spend planning where everything needs to go in your data center. And speaking of 3PAR storage, you can get all the performance you need for your most latency-sensitive applications as well as the data protection and availability options needed to mitigate risk. HPE Synergy Composer is a management appliance that uses software-defined intelligence with embedded HPE OneView to aggregate compute, storage, and fabric resources in a manner that scales to the needs of the application. With HPE OneView Rest API and automation tools, the entire process of server personality definition and configuration can be automated. The HPE Synergy Image Streamer management appliance works with the HPE Synergy Composer for fast, software-defined control over physical compute modules with operating system and application provisioning. HPE Synergy Composer powered by HPE OneView captures the physical state of the server in a server profile.

HPE Synergy HPE introduced new composable advancements to increase performance, improve security, and extend storage options for HPE Synergy, including: Faster storage provisioning—HPE Synergy customers will receive new SDS capabilities that allow IT administrators to easily and quickly compose the industry’s most complete offering of composable storage choices, including SAN, DAS, and SDS using HPE Synergy Composer and HPE OneView templates. Improved security—HPE has revealed the world’s most secure industry-standard

Technet24

||||||||||||||||||||

||||||||||||||||||||

servers with the introduction of its next-generation HPE ProLiant and HPE Synergy portfolio. HPE is the first vendor to put silicon-based security into its industry-standard platforms, addressing firmware attacks, which are one of the biggest threats facing enterprises and governments today. Increased performance—New HPE Synergy 480 and 660 Gen10 compute modules deliver increased performance across compute- and data-intensive workloads, such as financial modeling. HPE Synergy also delivers breakthrough Ethernet 25/50Gb connectivity with 100Gb uplinks, and a 25% increase in DAS capacity. HPE Synergy is the world’s fastest virtualization server backed by its record-breaking VMmark 2.x benchmark result. The HPE Synergy 12000 Frame is the hardware foundation for HPE Composable vision and provides a location in which compute, storage, fabric, and management are all aggregated. The HPE Synergy 12000 Frame is optimized to include all elements to run any workload. It includes embedded management and is designed to support a wide range of compute modules and storage options. The back of the frame includes fabric interconnects, enabling flexibility with regard to storage and other interconnectivity needs.

Virtual resource pools for any application and data type The D3940 storage module is designed for use in Synergy 12000 frames. Using the HPE Synergy 12Gb SAS connection module, it provides composable DAS for up to 10 compute modules in a single frame. Synergy storage is optimized for use as either a DAS array or as SDS using HPE StoreVirtual VSA or similar solutions. This 40 SFF (small form factor) drive bay module enables the creation of logical drives for any compute module in the Synergy frame. Each D3940 storage module supports between 8 and 40 hot-plug SFF SSDs using SAS and SATA (Serial ATA). Each Synergy frame supports up to five storage modules, thus enabling a single frame to deliver up to 160 disk drives total. D3940 storage modules feature simple configuration and setup with the Composer. They are easy to maintain and troubleshoot with industry-standard tools. These modules use a high-performance SAS connection with sixteen 12Gb/s SAS lanes. With this design, Synergy storage can deliver up to eight times the bandwidth when compared to JBODs (Just a Bunch Of Disks) and provide up to 12 Mb IOPS. It supports single or dual 12Gb IO modules, with dual modules providing nondisruptive updates. Various storage options for HPE Synergy are shown in Figure 3-48.

||||||||||||||||||||

||||||||||||||||||||

Note For more information, visit QuickSpecs: https://www.hpe.com/h20195/v2/GetDocument.aspx?docname=c04815113

Figure 3-48 Storage in HPE Synergy

Technet24

||||||||||||||||||||

Activity: Using HPE Reference Architecture Instructions: Use the “HPE Reference Architecture for Microsoft SQL Server 2014 on HPE SimpliVity 380” at: http://h20195.www2.hpe.com/V2/GetDocument.aspx? docname=a00021464enw for this activity. Answer the following questions: –

Which system is being used for this test?

–

List the bill of materials for a small enterprise solution?

–

What was the suggested RPO/RTO (Recovery Point Objective/Recovery Time Objective)?

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Learning check What components are included within the hyperconverged systems?

Write a summary of the key concepts presented in this chapter.

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Summary Virtualization is the pooling of multiple network storage devices into what appears to be a single storage device that is managed from a central console. HPE Virtual Connect for BladeSystem provides simplified storage, server, and network management in virtualized environments. HPE SAN Fabric topologies support a range of high-availability solutions, including interoperability in heterogeneous networks. With HPE OneView, you can view and manage your storage system and storage pools, add existing volumes, create new volumes, and create volume templates. HPE Hyper Converged system is an entire data center in a box. HPE Composable Infrastructure is designed around fluid resource pools, softwaredefined intelligence, and a unified API.

Technet24

||||||||||||||||||||

||||||||||||||||||||

4

HPE Storage Services

||||||||||||||||||||

LEARNING OBJECTIVES After completing this chapter, you should be able to identify and describe the services and support offered by Hewlett Packard Enterprise (HPE) Services, including: HPE Partner Ready Services Packaged service levels, including: –

HPE Foundation Care

–

HPE Proactive Care

–

HPE Proactive Care Advanced

–

HPE Datacenter Care

||||||||||||||||||||

HPE Solution Consulting Services HPE Financial Services (HPEFS) HPE Pointnext Services

Important HPE services vary by region. Consult your local HPE services representative for more information regarding local services.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Prelearning check In the HPE Partner Ready Program, there are six competencies. List them below.

||||||||||||||||||||

||||||||||||||||||||

Partner Ready Services The HPE Partner Ready for Service Providers Program allows more choice over previous years, including new electives tied to financial benefits to improve top-line revenue and bottom-line results for partners. The new structure is designed to allow partners to select one or more electives that meet their business requirements. Enhancements to the program include: Expanded HPE portfolio to permit an easier path to meet requirements criteria. Planned market development funds (MDFs) for service providers. New electives to enhance partner differentiation. HPEFS Flexible Capacity. Partner training and education. The Partner Ready Delivery Services program, shown in Figure 4-1, gives high-growth storage partners the flexibility to design services specifically for their customers and minimize service delivery risk by taking advantage of services and tools. Depending on their investment in their services infrastructure, partners can resell HPE services, sell and deliver HPE services, or sell their own brand of services over the customer’s solution lifecycle. Partners can also create combinations of these options that optimize the service opportunity. HPE is the only vendor to have both a sales model and a selland-deliver model for services. This program can help organizations offer services that target new markets or combine services into an integrated partner-branded service offering. Note Market development funds or MDFs are used in an indirect sales channel where funds are made available by a manufacturer or brand to help affiliates, channel partners, resellers, value-added resellers (VARs), or distributors, etc. sell its products and create local awareness about the national brand. (April 2018, Wikipedia.org)

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 4-1 Partner Ready Program

Value in partnership The Partner Ready Delivery Services program gives high-growth storage partners the flexibility to design services specifically for their customers and minimize service delivery risk by taking advantage of services and tools. Depending on their investment in their services infrastructure, partners can resell HPE services, sell and deliver HPE services, or sell their own brand of services over the customer’s solution lifecycle. Value of Partner Ready Delivery Services program is shown in Figure 4-2.

Figure 4-2 Value in partnership

||||||||||||||||||||

||||||||||||||||||||

Business benefits of the Partner Ready Services As cloud computing is now regarded as mainstream technology, HPE partner ready for service provider partners have new opportunities that move beyond their traditional infrastructure hosting business to offer managed private clouds, virtual private clouds, and public clouds. The HPE strategy for service providers is based on three pillars, shown in Figure 4-3, to enable partners to grow revenue streams through solutions, adapt financial offerings that best align operational costs to revenue, and collaborate with HPE to drive sales execution.

Figure 4-3 Business benefits of the Partner Ready Services The key HPE differentiators include: Multiple avenues for partners to extend their reach, enhance their capabilities, and expand their business and profit. 24x7 hardware and software support, escalation management, installation advisory support, two levels of replacement parts service, and access to electronic support and services. License to use (LTU) software updates, and software product and documentation updates. Automated support tools, such as HPE Insight Remote Support Advanced, that provide responsive, efficient, and effective services. Integrated hardware, software, and solution support.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Earning competencies Competencies, shown in Figure 4-4, are a new component of the Partner Ready Program that recognize and enable partners who have specific capabilities in targeted solutions.

Figure 4-4 Earning competencies in Partner Ready Program Participation requirements and procedure Compile proof of successful customer deployments (3). Complete Competency curriculum. Contact partner business manager. Submit customer references. Communicate with HPE as HPE validates references. Six Competencies are available: Partner has application assessment, architecture, and development experience with both traditional and cloud-based applications. Provides solutions to migrate applications to cloud. Can assist customers with re-hosting or re-architecting their applications to leverage scalable cloud platforms. Partner has a detailed understanding of cloud management and automation; core infrastructure including server, storage, networking, and integrations; and virtualization. Provides rapid provisioning, data center automation, and software-defined infrastructure solutions. Can automate and simplify deployment across cloud environments. Partner has experience with architecting secure storage solutions that leverage multiple tiers and data management solutions, specifically HPE Backup & Recovery. Partners with advisory capabilities will provide consulting on BC/DR (business continuity and disaster recovery) planning and recovery exercises. Partner has significant knowledge and understanding of data provisioning and online

||||||||||||||||||||

||||||||||||||||||||

transaction processing (OLTP). Must have experience with large volumes of streaming data and understand tools and processes to ensure data quality. Should have understanding of architecture required to support DBaaS (Database-as-a-service). Provides HPE services, support, and infrastructure to maximize adoption of SAP HANA. Partner has expertise in desktop operating systems and applications. Provides secure virtual workstation solutions. Can offer server, storage, networking, and application expertise in supporting multitenant remote client installations including remote display, virtual application, and full virtual desktops. Partner has detailed understanding of enterprise communications applications, including Skype for Business, in both a public and private cloud configuration. Communication applications include secure messaging, video conferencing, and voice services.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Accessing the Partner Ready Portal Instructions: Log in to the HPE Partner Ready Portal at: https://partner.hpe.com/login for this activity. List the six competencies required for the FY18 program.

||||||||||||||||||||

||||||||||||||||||||

Learning check In the HPE Partner Ready Program, there are six competencies. List them below.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Overview of HPE Storage support services Many storage environments become more complex as capacity increases. With equipment and contracts from multiple vendors, customers are likely losing time and money by having to repeatedly integrate components and systems. Storage technology services that combine technical expertise with multivendor expertise can help businesses operate more efficiently and strategically and deliver measurable return on investment. To improve efficiency, performance, and continuity in today’s highly integrated and virtualized environments, businesses need a support vendor who can look outside the storage device to identify the root of problems before they spread. HPE experts can help simplify multivendor storage environments. HPE expertise spans the entire technology environment, not just storage. With storage consulting and support services, customers can be confident that their storage needs are met. HPE expertise spans storage infrastructure transformation, data optimization and availability, storage service transformation, multivendor storage area network (SAN) support, installation, and implementation. HPE offers the industry’s broadest portfolio of storage services, so customers can choose levels of reactive and proactive support. From mission-critical onsite services to innovative web-based remote support, customers can choose the precise level of attention their business demands. With service experts available worldwide, HPE supports every phase of the storage lifecycle. HPE Lifecycle Event Services are available directly from HPE and through a global network of authorized Support Partners. From a simple installation to ongoing assistance to augment a customer’s onsite technical staff, HPE expertise enables customers to meet the needs of their business. HPE OneView Startup Installation and Configuration Service is a basic fixed-price, fixed-scope installation and startup service that includes the installation and configuration of a single HPE OneView virtual appliance on a supported hypervisor host (VMware® vSphere or Microsoft® Hyper-V) on a supported HPE ProLiant server (blade or rack environment). In a blade environment, it provides configuration and setup of all managed devices within a single HPE BladeSystem c7000 enclosure. In a rackmount environment, it provides configuration and setup of all managed devices within a single HPE ProLiant server rack. HPE 3PAR Health Check Service provides a scorecard assessment of system configuration, capacity, and supportability for a single HPE 3PAR Storage System. The assessment results will help you to improve the utilization of your HPE 3PAR Storage

||||||||||||||||||||

||||||||||||||||||||

System and identify areas of potential concern before they impact availability or performance. The assessment information is compared to HPE best practices, and HPE provides your organization with a summary report that includes recommendations to help you improve the availability and ongoing management of your HPE 3PAR Storage System. The HPE specialist will examine several parameters, including whether the host I/O is within the node limits, that all hosts have at least two paths to the array, that the data is appropriately balanced over the disks in the array, and that there is adequate free capacity, as well as other key configuration and performance indicators. Installation Service provides for the basic installation of HPE branded equipment, software products, as well as HPE-supported products from other vendors that are sold by HPE or by HPE authorized resellers. The Installation Service is part of a suite of HPE deployment services that are designed to give you the peace of mind that comes from knowing your HPE and HPE-supported products have been installed by an HPE specialist in accordance with the manufacturer’s product documentation.

Support Services Central Support Services Central, shown in Figure 4-5, formerly Care Pack Central gives you to quickly identify Packaged Support Services. This site includes StoreVirtual, fabric, and interconnect devices along with tape storage. This site does not include Enterprise storage support, for Enterprise storage products, visit: http://ssc.hpe.com/portal/site/ssc/?action=determineNodeContents&nodeId=28965

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 4-5 Support Services Central

HPE Support Center access Foundation Care, Proactive Care, and Proactive Care Advanced customers have access to the HPE Support Center. The HPE Support Center provides access to a personalized Insight Online dashboard, support forums, support case submittal, drivers, patch management, software updates, and warranty/contract coverage. HPE Support Center access and functionality are enabled through the linking of the customer’s HPE Passport with service agreements. The service agreements must be linked to the HPE Passport account to enable all available features. The entire HPE Support Center portal experience is available as part of your HPE standard warranty, HPE packaged services, or contractual support agreement with HPE. You have the flexibility and control to get the services you need, when you need them, each year. Our new and enhanced support menu offers a selection of around 100 IT services for the digital age.

||||||||||||||||||||

||||||||||||||||||||

Just a few of the offerings include: Blade Toolset Assessment Virtualization Performance Audit Storage Firmware and Software Analysis and Management SAN Firmware and Software Analysis and Management Services for ConvergedSystem for SAP HANA® Network Firmware and Software Analysis

Firmware updates for eligible products Firmware updates for HPE hardware products are available only to customers with an active agreement that entitles them to access these updates. Foundation Care, Proactive Care, and Proactive Care Advanced customers have the right to download, install, and use firmware updates for hardware products covered by this service, subject to all applicable license restrictions. HPE will verify entitlement to updates by reasonable means (such as an access code or other identifier), and the customer is responsible for using any such access tools in accordance with the terms of applicable agreements with HPE. HPE might take additional steps, including audits, to verify the customer’s adherence to terms of their agreements with HPE. If a customer has licenses to features implemented in firmware activated by the purchase of a separate software license product, they must also have, if available, an active HPE Software Support agreement to receive, download, install, and use related firmware updates.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Auditing and call-to-repair times If an upfront audit is required by HPE, the hardware call-to-repair time will not take effect until five business days after the audit has been completed. In addition, HPE reserves the right to downgrade service to an onsite response time or cancel the service contract if critical audit suggestions are not followed or the audit is not performed within the specified time frame. The hardware repair time might vary for specific products. A call-to-repair time does not apply when the customer chooses to have HPE prolong diagnosis rather than execute recommended server recovery procedures. If the customer requests scheduled service, the repair time frame begins from the agreed-upon scheduled time. HPE reserves the right to modify the call-to-repair time as it applies to the customer’s specific product configuration, location, and environment. These activities or situations suspend the call-to-repair time calculation until they are completed or resolved: Any customer or third-party action or inaction impacting the repair process Any automated recovery processes triggered by the hardware malfunction, such as disk mechanism rebuild or sparing procedures Any other activities not specific to the hardware repair that are required to verify that the hardware malfunction has been corrected, such as rebooting the operating system

Customer self-repair Customer self-repair (CSR) is a key component of HPE warranty terms. It allows HPE to ship replacement parts directly to customers after a failure has been confirmed. Parts are generally shipped overnight, so they are received as quickly as possible. Customers can then replace the parts at their convenience. HPE has invested significantly in efforts to engineer products to be customer repairable. CSR offers the most convenience in HPE product repairs. Generally, it is the fastest way to implement a repair. Customers do not have to wait for a service engineer. They can control when the repair is implemented to accommodate their routines and resources. CSR-enabled products help HPE reduce the overall cost of the warranty and services, providing a lower total cost of ownership. CSR is the required method of repair when a product is classified as “mandatory CSR,” and CSR is offered within the country. For mandatory CSR products, replacement parts are classified as follows:

||||||||||||||||||||

||||||||||||||||||||

CSR mandatory parts (warranty only)—These parts require customer replacement. They typically have toolless access, consist of a single assembly part, require minimal cabling, and have plug-and-play configuration. CSR optional parts—Customers can optionally elect to replace these parts. They might require tools, the removal of other parts, more involved cabling, or potential configuration/setup following replacement. Non-CSR—Non-CSR parts do not have specific CSR classification or instructions on customer replacement. Note More information about CSR: https://www.hpe.com/support/sml-csr

Note The Partner Services Media Library features animations and videos illustrating service and support procedures for many HPE products enabled for partner support. This just-in-time training tool allows you to quickly identify and access information needed to assist with teardown, part replacement, assembly, and other common hardware procedures. Important! Use of the Partner Services Media Library by individuals who are not authorized under partner programs is prohibited. http://www.hpe.com/support/psml

Replacement parts For replacement parts and components that are discontinued, an upgrade path might be required. Upgrades for discontinued parts or components might result in additional charges to the customer. HPE will work with the customer to recommend a replacement. Not all components have available replacements in all countries because of local support capabilities. Consumable items including, but not limited to, removable media, customer-replaceable batteries, maintenance kits, and other supplies, as well as user maintenance, are not covered by this service. Some servers, networking, and storage products; CPUs; disks; and other major internal and external components are covered if support has been configured accordingly and they are listed in the contract’s equipment list under the hardware support section (if applicable). For ProLiant servers, networking, and storage systems, the service on the main product

Technet24

||||||||||||||||||||

||||||||||||||||||||

covers HPE hardware options that do not require separate coverage, are qualified for the server, are purchased at the same time or afterward, and are internal to the enclosure. These items will be covered at the same service level as the main product. For ProLiant servers or storage or Blade System enclosures installed within a rack, the service also covers all HPE-qualified rack options not designated by HPE as requiring separate coverage. The UPS battery is covered separately under its own warranty terms and conditions and is limited to the terms of the applicable warranty period.

Package-specific replacement part terms For Foundation Care, Proactive Care, and Proactive Care Advanced customers, HPE provides replacement parts and materials necessary to keep covered hardware products in operating condition, including parts and materials for available and recommended engineering improvements. Replacement parts provided by HPE will be new or functionally equivalent to new in performance. All replaced parts become the property of HPE unless optional defective material retention or comprehensive defective material retention (CDMR) service options have been purchased. Customers without a defective media retention (DMR) or CDMR contract who wish to retain, degauss, or otherwise physically destroy replaced parts are billed for the list price of the replacement part. Supplies and consumable parts are not supported and are not provided as part of this service. Standard warranty terms and conditions apply to supplies and consumable parts. The repair or replacement of any supplies or consumable parts is the responsibility of the customer. Some exceptions might apply. Customers can contact HPE for more information. If a consumable part is eligible for coverage, as determined by HPE, call-to-repair time commitments and onsite response times do not apply to repair or replacement of the covered consumable part. Note Parts and components that have reached their maximum supported lifetime or the maximum usage limit as outlined in the manufacturer’s operating manual, product QuickSpecs, or technical product data sheet are not provided, repaired, or replaced as part of this service.

Severity levels Hardware support onsite, call-to-repair, and software support remote response times

||||||||||||||||||||

||||||||||||||||||||

differ depending on incident severity and the purchased onsite coverage level. The customer determines the incident severity level when logging or confirming a case with HPE. Note More information about HPE Hardware Support Onsite Service and HPE Packaged Support Services is provided in: https://h20195.www2.hpe.com/v2/GetPDF.aspx/5982-6547ENN.pdf

Technet24

||||||||||||||||||||

Activity: Understanding severity levels: Part 1 Instructions: Use the “HPE Hardware Support Onsite Service” documentation at: https://h20195.www2.hpe.com/V2/GetPDF.aspx/5982-6547EN.pdf for this activity. Write down the definition of each of the four severity levels: –

Severity 1:

–

Severity 2:

–

Severity 3:

–

Severity 4:

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Activity: Understanding severity levels: Part 2 Instructions: Match each severity level with its definition in Figure 4-6.

Figure 4-6 Understanding severity levels: Part 2

Travel zones

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 4-7 Travel zones All hardware onsite response times apply only to sites located within 25 miles (40 kilometers) of an HPE support hub. Travel to sites located within 200 miles (320 kilometers) of an HPE support hub is provided at no additional charge. There is an additional travel charge if the site is located more than 200 miles (320 kilometers) from the HPE support hub. Travel zones and charges might vary in some geographic locations. Response times to sites located more than 100 miles (160 kilometers) from an HPE support hub will be modified for extended travel, as shown in Figure 4-8. Travel zones and charges, if applicable, might vary in some geographic locations.

Figure 4-8 Travel zones A hardware call-to-repair time is available for sites located within 50 miles (80 kilometers) of an HPE support hub. Travel zones and charges might vary in some geographic locations. The hardware call-to-repair time is not available for sites located more than 100 miles (160 kilometers) from an HPE support hub. For sites located between 51 and 100 miles

||||||||||||||||||||

||||||||||||||||||||

(82 to 160 kilometers) of an HPE support hub, an adjusted hardware call-to-repair time applies, as shown in Figure 4-9.

Figure 4-9 Packaged service levels

Customer responsibilities If the customer does not act upon their specified responsibilities, HPE or the HPE authorized service provider will not be obligated to deliver the services as described or will perform the service at the customer’s expense at the prevailing time and material rates. The customer must provide accurate and complete information in a timely manner as required for HPE to perform the services. HPE can request that the customer supports its remote problem resolution efforts. The customer might be required to: Start self-tests and install and run other diagnostic tools and programs. Install software and firmware updates and patches. Run data collection scripts on behalf of HPE when they cannot be initiated from Remote Support Technology. Provide all information necessary for HPE to deliver timely and professional remote support and determine the level of support eligibility. Perform other reasonable activities to help HPE identify or resolve problems, as requested.

Technet24

||||||||||||||||||||

The customer will take responsibility for registering to use the HPE or a third-party vendor’s electronic facility to access knowledge databases and obtain product information. HPE will provide registration information to the customer as required. Additionally, for certain storage products, the customer might be required to accept vendor-specific terms for use of the electronic facility. The customer will: Retain and provide to HPE all original software licenses, license agreements, license keys, and subscription service registration information on request. Take responsibility for acting upon software product updates and obsolescence notifications received from the HPE Support Center. Use all software products in accordance with current HPE software licensing terms corresponding to the customer’s prerequisite underlying software license or in accordance with the current licensing terms of the third-party software manufacturer. If the customer chooses to retain repair parts covered under the DMR and CDMR service feature options, it is the customer’s responsibility to: Retain physical control of the covered data-retentive components at all times during support delivery by HPE. Ensure that any customer-sensitive data on the data-retentive component is destroyed or remains secure. Have an authorized representative present to retain the defective data-retentive component, accept the replacement component, provide HPE with identification information for each data-retentive component, and complete a document provided by HPE acknowledging the retention of the data-retentive component. –

Destroy the data-retentive component and ensure that it is not used again.

–

Dispose of all data-retentive components in compliance with applicable environmental laws and regulations.

||||||||||||||||||||

Service limitations HPE retains the right to determine the final resolution of all service requests. At the discretion of HPE, service will be provided using a combination of remote and onsite diagnosis and support. Other service delivery methods might include shipped delivery of replaceable parts such as a keyboard, mouse, certain hard disk drives, and other parts classified by HPE as CSR parts. HPE will determine the appropriate delivery method required to provide effective and timely support and meet the call-to-repair time, if applicable. If the customer agrees to the recommended CSR and a CSR part is provided to return

||||||||||||||||||||

||||||||||||||||||||

the system to operating condition, onsite service is not required. In such cases, HPE express ships the CSR parts that are critical to operations. Call-to-repair and onsite response times do not apply to the repair or replacement of defective or depleted batteries for selected enterprise storage arrays and enterprise tape products. Call-to-repair and onsite response times also do not apply to activities such as: Services required due to failure of the customer to incorporate any system fix, repair, patch, or modification provided by HPE. Services that are required due to unauthorized attempts by third-party personnel to install, repair, maintain, or modify hardware, firmware, or software. Operational testing of applications or additional tests requested or required by the customer. Services that are required due to improper treatment or use of the products or equipment. Services required due to failure of the customer to take preventive action previously advised by HPE. Backup and recovery of the operating system, other software, and data. Installation of any customer-installable firmware and software updates. Troubleshooting for interconnectivity or compatibility problems. Support for network-related problems.

Packaged service levels HPE packaged service levels, shown in Figure 4-9, include: Foundation Care—With HPE Foundation Care, customers can minimize the time spent on troubleshooting, monitoring, and remediating while maintaining availability within limited budgets and resources. Proactive Care and Proactive Care Advanced—This simplified and cost-effective option combines high-level personal service with smart technology, all to maintain an agile, healthy, and reliable infrastructure that boosts business performance. Remote automated consultation, recommendations, and support deliver fast troubleshooting and problem resolution. Datacenter Care—Customers get flexible, relationship-based support and management

Technet24

||||||||||||||||||||

||||||||||||||||||||

for their heterogeneous data centers. A single point of contact and dedicated account team helps improve stability in a current IT environment and provides the expertise on particular technologies to guide customers further to meet their goals. Certain support levels and services are standard across all packaged service levels, while others are specific to certain package levels. Note 1.  Post-incident reviews are at the discretion of HPE. 2.  A device indicates a configured device such as a chassis configured with components and software.

HPE Foundation Care Foundation Care includes comprehensive hardware and software services aimed to help increase the availability of customers’ IT infrastructure. HPE technical resources work with IT teams to help resolve hardware and software problems with HPE and selected third-party products. HPE experts can help diagnose problems, access firmware, or provide software updates and replacement parts. For hardware products covered by Foundation Care, the service includes remote diagnosis and support, as well as onsite hardware repair when needed. For eligible HPE hardware products, this service might also include Basic Software Support and Collaborative Call Management for selected third-party software. For software products covered by Foundation Care, HPE provides remote technical support and access to software updates and patches. HPE releases updates to software and reference manuals for selected software products that are compatible with HPE products. These updates cover systems, processors, processor cores, and users, as allowed by HPE or the original manufacturer software license. Updates for selected third-party software products are included as they are made available from the original software manufacturer. In addition, Foundation Care provides electronic access to related product and support information, enabling any member of a customer’s IT staff to locate this commercially available information. For third-party products, access is subject to availability of information from the original manufacturer. Customers can choose from a set of reactive support levels to meet business and operational needs. Foundation Care provides support that is simple, affordable, scalable, and personal. It helps customers resolve problems faster to keep the business

||||||||||||||||||||

||||||||||||||||||||

running. The HPE network includes more than 70,000 channel partners worldwide and partnerships with leading industry vendors. This provides support for hardware and software across the IT infrastructure, including qualified experts certified in multivendor technologies. Overview of basic, standard, and optimized foundation care levels is shown in Figure 4-10.

Figure 4-10 HPE Foundation Care

HPE Foundation Care support levels Next Business Day For Foundation Care Next Business Day hardware and software support, service is available nine hours per day, between 8:00 a.m. and 5:00 p.m. in the customer’s time zone, Monday through Friday, excluding holidays observed by HPE. Hardware support includes onsite response time. For incidents with covered hardware that cannot be resolved remotely, HPE will use commercially reasonable efforts to respond onsite the next business day. An HPE authorized representative will arrive at the customer’s site during the coverage window to begin hardware maintenance service the next coverage day after the call has been received and acknowledged by HPE. Note Onsite response time begins when the initial call has been received and acknowledged by HPE. The onsite response time ends when the HPE authorized representative arrives at the customer's site, or when the reported event is closed after HPE has determined that it does not require an onsite intervention. Calls received outside the coverage window will be acknowledged the next coverage day and serviced within the following coverage day.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Software support includes remote response time. After a software problem is logged, an HPE Solution Center engineer will respond to the call within two hours.

Foundation Care 24x7 For Foundation Care 24x7 hardware and software support, service is available 24 hours per day, seven days per week including holidays observed by HPE. Hardware support includes four-hour onsite response for incidents with covered hardware that cannot be resolved remotely. HPE will use commercially reasonable efforts to respond onsite within four hours. An HPE authorized representative will arrive at the customer’s site during the coverage window to begin hardware maintenance service within four hours of the call having been received and acknowledged by HPE. Foundation Care 24x7 software support includes remote response time. After a software problem is logged, an HPE Solution Center engineer will respond to the call within two hours.

Foundation Care call to repair To support HPE call-to-repair times, an inventory of critical replacement parts is maintained for customers who have selected the call-to-repair option. This inventory is stored at an HPE facility. These parts are accessible to HPE authorized representatives responding to eligible service requests. Foundation Care Call-to-Repair hardware support includes: Coverage window—Service is available 24 hours per day, seven days per week including holidays observed by HPE. Six-hour call-to-repair time—For critical incidents (severity 1 or 2), HPE will use commercially reasonable efforts to return the covered hardware to operating condition within six hours after the initial service request has been received. Availability of callto-repair times depends on the proximity of the customer site to an HPE support hub. For noncritical incidents (severity 3 or 4) or at the customer’s request, HPE will schedule a time, and the call-to-repair time will then start. HPE is not liable for any lost data, and the customer is responsible for implementing appropriate backup procedures. To verify that the work is complete, HPE might perform a power-on self-test (POST), stand-alone diagnostic test, or visual verification of proper operation. At its sole discretion, HPE will determine the level of testing necessary to verify that the hardware is repaired.

||||||||||||||||||||

||||||||||||||||||||

HPE might temporarily or permanently replace the product to meet the call-to-repair time. Replacement products are new or functionally equivalent to new in performance. Replaced products become the property of HPE. It will take 30 days from the time this service is purchased to set up and perform necessary audits and processes so that the hardware call-to-repair time can be put in effect. During this initial 30-day period and for up to five additional business days after the audit is complete, HPE will provide a four-hour onsite response time.

Additional service levels The Foundation Care portfolio also offers the following additional service levels: Foundation Care Next Business Day with DMR Service Foundation Care 24x7 with DMR Service Foundation Care Call to Repair with DMR Service These optional service features include hardware DMR and CDMR. With these features, customers can retain sensitive data within a defective hard disk or eligible solid-state drive (SSD)/flash drive. All disks or eligible SSD/flash drives on a covered system must participate in the DMR service option. This feature allows the customer to retain defective hard disk or eligible SSD/flash drive components that the customer does not want to relinquish because of sensitive data contained within the disk. The CDMR service feature option allows the customer to retain additional components that HPE designates as having data-retentive capabilities, such as memory modules. All eligible data-retentive components on a covered system must participate in the CDMR service option. After a noncritical software service request (of severity level 3 or 4) is received and logged, HPE will respond to the call within two hours. HPE provides corrective support to resolve identifiable and customer-reproducible software product problems. HPE also provides support to help the customer identify problems that are difficult to reproduce. The customer receives assistance with troubleshooting incidents and resolving configuration parameters.

HPE Proactive Care Proactive Care, benefits shown in Figure 4-11, offers an integrated set of reactive and proactive services designed to help improve the stability and operation of a converged infrastructure. In a complex converged and virtualized environment, many components

Technet24

||||||||||||||||||||

||||||||||||||||||||

need to work together effectively. Proactive Care is designed to support devices in these environments, covering servers, operating systems, hypervisors, storage, SANs, and networks. Proactive Care Services include consultation, recommendations, and onsite hardware repair when needed.

Figure 4-11 HPE Proactive Care Customers can choose from a range of hardware reactive support levels to meet business and operational needs. In the event of a service incident, Proactive Care provides access to advanced technical solution specialists, who will manage that case from start to finish. HPE uses enhanced incident management procedures intended to provide rapid resolution of complex incidents. Additionally, technical solution specialists are equipped with automation technologies and tools designed to help reduce downtime and increase productivity. Proactive Care Services include firmware and software version analysis for supported devices and a list of recommendations to maintain recommended revision levels. Customers receive a regular proactive scan of devices covered by Proactive Care, which can help identify and resolve configuration problems. Proactive Care Services use HPE proprietary service tools, which are referred to as Remote Support Technology, to enable faster delivery of services by collecting technical configuration and fault data. This is the principal method for delivering device monitoring, automated case creation, and a variety of proactive reports. The current version of Remote Support Technology, with the data collections function enabled, is a prerequisite for delivery of Proactive Care Services; the customer is responsible for installing the software. If customers do not install and operate the current version of the software, HPE will not provide the firmware and software version report, proactive scan report, hardware call-to-repair time commitment, remote monitoring, or automated call logging. To ensure a successful installation, HPE will provide remote technical advice and assistance on the installation and configuration on the customer’s request. HPE will

||||||||||||||||||||

recommend the appropriate configuration based on the type and number of devices supported in the customer’s Proactive Care environment.

Hardware and software incident support Each Proactive Care level includes problem prevention and incident management support for hardware and software products. For each Proactive Care level, HPE provides all the core problem prevention services and the related core incident management service features. For hardware products, the Proactive Care portfolio offers three distinct hardware service levels: Next Business Day Proactive Care Service –

Hardware support coverage window—Onsite service is available between 8:00 a.m. and 5:00 p.m. within the customer’s time zone, Monday through Friday, excluding holidays observed by HPE.

–

Hardware onsite support response time—An HPE authorized representative will arrive at the customer’s site during the onsite coverage window to begin hardware maintenance service on the next coverage day after the call has been received and acknowledged by HPE.

Four-hour 24x7 Proactive Care Service –

Hardware support coverage window—Service is available 24 hours per day, seven days per week including holidays observed by HPE.

–

Hardware onsite support response time—An HPE authorized representative will arrive at the customer’s site to begin hardware maintenance service within four hours after the call has been received and acknowledged by HPE.

Six-hour Call-to-Repair Proactive Care Service –

Hardware support coverage window—Service is available 24 hours per day, seven days per week including holidays observed by HPE.

–

Hardware call-to-repair time commitment—For critical incidents (severity 1 and 2), HPE will use commercially reasonable efforts to return the covered hardware to operating condition within six hours after the call has been received and acknowledged by HPE. Availability of response times and call-to-repair times depends on the proximity of the customer site to an HPE support hub. For noncritical incidents (severity 3 and 4) or at the customer’s request, HPE will schedule a time for the remedial action to begin, and the call-to-repair time commitment will start at that time.

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

To support HPE call-to-repair time commitments, an inventory of critical replacement parts is maintained for customers who have selected the call-to-repair option. This inventory is stored at an HPE facility. These parts are managed to allow for increased inventory availability and are accessible to HPE representatives responding to eligible calls. Note Call-to-repair time refers to the period of time that begins when the initial hardware service request has been received and acknowledged by HPE or when the work begins. Call-to-repair time ends when the repaired is complete or the reported event is closed with the explanation that it does not currently require onsite intervention.

HPE Proactive Care Advanced Proactive Care Advanced services expand on Proactive Care Services. They are designed to help customers maximize the benefits of IT investments, maintain IT infrastructure stability, achieve business and IT project objectives, reduce operational costs, and free IT staff to perform high-priority tasks. Proactive Care Advanced can help customers save time with real-time monitoring and analysis of devices that are connected to HPE, creating personalized proactive reports with recommendations to help prevent problems within a customer IT infrastructure. Proactive Care Advanced provides: Faster incident resolution from specially trained, solution-oriented, advanced HPE specialists who manage the case from start to finish. Recommendations for firmware version and software patching on supported products to help prevent problems. Identification of risks and issues through regular device-based proactive scans that help ensure that configurations are consistent with HPE best practices. Access to technical advice and services from HPE specialists to complement customers’ existing IT resources. Proactive Care Advanced assigns each customer an HPE Account Support Manager (ASM) who provides personalized technical and operational advice, including best practices from HPE. The assigned ASM can arrange specialist technical advice to assist with specific projects, performance improvements, or other technical needs.

||||||||||||||||||||

||||||||||||||||||||

An HPE Technical Support Specialist (TSS) delivers an enhanced call experience intended to provide fast incident resolution. For severity 1 incidents, a Critical Event Manager (CEM) is accountable for the case and provides regular status updates. This advanced program uses Remote Support Technology to monitor devices and collect data, enabling faster delivery of support and services. Running the current version of Remote Support Technology is required to receive full delivery and benefits from this support service. Proactive Care Advanced includes credits that customers can use to select and fund any specialized service assistance. Customers can choose from a range of predefined technical services on the HPE Proactive Select menu, or the designated ASM can work with customers to define the specific advice or assistance that is needed.

HPE Lifecycle Event Services For smooth startup, the HPE 3PAR Storage System Installation and Startup Service provides deployment of your HPE 3PAR Storage System. Helping to ensure proper installation in your storage environment—and helping you realize the maximum benefit from your storage investment. The service provides the necessary activities required to deploy your HPE 3PAR Storage System into operation. With the assistance of your designated IT storage administrator, a service specialist deploys your array. When ordered with the upgrade products, the service also provides deployment of hardware upgrades to your existing HPE 3PAR Storage System. The service includes: For new arrays, configuration and presentation of Virtual Volumes for up to two hosts. For array upgrades, installation and configuration of the array upgrade products and the minimum activities required to present the upgrade products to designated hosts. For array rack transition, removal of the array from the HPE 3PAR factory rack and installation of the array into a rack you supply that meets HPE requirements. Reconfiguration of your existing array—for example, Virtual Volumes, hosts, or SAN— is excluded from this service. For installation of the rack version of the array into a rack you supply, assembly, configuration, and positioning of the rack are excluded from this service. Rack transition does not include installation of upgrades or relocation services. Rack Technet24

||||||||||||||||||||

||||||||||||||||||||

transition service delivery requires that the HPE 3PAR Storage System be temporarily powered down. Upgrade installation and relocation services are available separately. Additional configuration/integration activities can be accommodated at incremental cost. Note To access more information about this startup service, go to: https://www.hpe.com/h20195/v2/GetPDF.aspx/4AA2-8240ENW.pdf

HPE Insight Remote Support HPE Insight Remote Support is a software solution that enables reactive and proactive remote support to improve the availability of supported servers, storage, networking, and printers. HPE Insight Remote Support enables enhanced delivery of your warranty or support services contract. Notifications might be sent to your HPE Authorized Channel Partner for on-site service, if configured and available in your country. HPE Insight Remote Support 7.8 is optimized to support up to 3,500 monitored devices per Hosting Device and can be installed on a Windows ProLiant Hosting Device or a Windows Virtual Guest. Additionally, it can easily be configured to forward service events to a supported version of Systems Insight Manager and offers integration into Insight Online.

HPE ProLiant Gen10 servers Because Insight Remote Support (RS) monitors ProLiant Gen10 servers through the integrated Lights Out (iLO) 5, Insight RS supports any operating system that is also supported on the hardware. However, if you want Insight RS to gather operating system information, you must install Agentless Management Service (AMS) on the ProLiant Gen10 server for that particular operating system. HPE recommends that you read the documentation to assist during the installation. There are system prerequisites for both your Hosting Device and monitored devices that, if ignored, will prevent a successful installation and might require a full re-installation.

||||||||||||||||||||

||||||||||||||||||||

Note For prerequisites and installation instructions, see the HPE Insight Remote Support Quick Installation Guide available at http://www.hpe.com/info/insightremotesupport/docs

Enhanced call handling Enhanced call handling capabilities direct Proactive Care and Proactive Care Advanced customers to one of the HPE Centers of Excellence. Enhanced call handling is shown in Figure 4-12. As a result, clients receive the specific expertise and product support they need without being routed through multiple tiers of support. This provides rapid resolution for problems. Customers can also add features such as self-service spares. Call handing—The customer can contact HPE 24x7. When the customer calls with a critical incident, HPE aims to either connect the customer to a TSS or call the customer back within 15 minutes. If there is a hardware issue requiring onsite service, an HPE customer engineer is dispatched to the customer’s site in accordance with the purchased hardware onsite reactive service level for that affected device. After HPE has acknowledged and received the service request, HPE works to isolate the hardware or software problem and to troubleshoot, remedy, and attempt to resolve the problem remotely with the customer. HPE might initiate and perform remote diagnostic tests using innovative automation tools to access covered products. Automatic call logging capability—Incident cases for HPE products using Remote Support Technology can be automatically created 24x7. For supported devices, automatic call logging capabilities are enabled so that devices will submit hardware service incidents directly to HPE using Remote Support Technology. Where configured, HPE Insight Online can provide a single point of visibility to incidents and resolution. Basic Software Support and Collaborative Call Management—Basic Software Support provides 24x7 phone support for selected independent software vendor (ISV) software that resides on hardware covered by Proactive Care. With this support, HPE investigates and attempts to resolve problems by asking the customer to apply fixes that have been made available or known to HPE. Collaborative Call Management can be provided only when customers have appropriate active support agreements with selected ISVs, and the customer has taken the steps necessary to ensure that HPE can submit calls on the customer’s behalf to resolve the issue. HPE will engage the ISV and provide information about the customer’s issue.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Note Enhanced call handling capabilities are available for Proactive Care and Proactive Care Advanced customers only.

Figure 4-12 Enhanced call handling

HPE Datacenter Care Datacenter Care is the most comprehensive support solution from HPE. Tailored to meet customers’ specific data center support requirements, it offers a choice of proactive and reactive service levels to cover requirements ranging from the most basic to the most business-critical environments. Datacenter Care is designed to scale to any size and type of data center environment. Customers can use Datacenter Care, shown in Figure 4-13, to complement their existing skills and capabilities by mixing and matching any HPE support offerings with different elements of their IT solutions or data center. It reduces risk across people, processes, and technology; increases IT service quality and productivity; and reduces costs.

||||||||||||||||||||

||||||||||||||||||||

Figure 4-13 HPE Datacenter Care HPE will provide a single point of contact for HPE and select multivendor products. The service is delivered under the governance of an assigned HPE account team led by a trained ASM. The team’s goal is to form a close working relationship with designated members of customers’ IT staff and gain a clear understanding of business objectives, key service-level agreements, and key performance indicators. A mutually agreed-upon and executed statement of work (SOW) details the precise combination of reactive and proactive support features that will be provided under Datacenter Care based on business requirements. The service includes an assigned Delivery of the various support options will be overseen by the ASM and directed at meeting business goals.

Datacenter Care extensions When the unexpected happens, customers might still need rapid escalation and incident resolution. In the event of a service incident, Datacenter Care provides access to HPE technical solution specialists who can help customers resolve critical issues as quickly

Technet24

||||||||||||||||||||

as possible. HPE employees accelerate escalation procedures to resolve complex incidents. HPE specialists are equipped with remote technologies and tools designed to reduce downtime and increase productivity. A set of optional proactive services, including technology-specific activities such as firmware and operating system patch analysis/recommendations; change management support; a systematic approach to continual improvement based on IT service management; and HPE best practices, including IT Infrastructure Library and ISO/IEC 20000 standards. Note ISO/IEC 20000 is the first international standard for IT service management. It was developed in 2005, by ISO/IEC JTC1/SC7, and revised in 2011. It is based on and intended to supersede the earlier BS 15000 that was developed by BSI Group. (April 2018, Wikipedia.org) Datacenter Care also provides the following optional extensions: Flexible Capacity –

Flexible Capacity is an infrastructure utility service based on the converged infrastructure of HPE server, storage, and networking equipment installed at customer sites that are billed based on usage. This allows businesses to procure and pay for capacity needs on a variable monthly usage basis, subject to minimum usage requirements.

Datacenter Care Operational Support Services –

||||||||||||||||||||

Operational Support Services provide HPE best practices for operating on-premise infrastructure, including 24x7 remote infrastructure monitoring and operational services. It supports the service operations stage of the infrastructure, servers, storage, networking, operating system, hypervisor, backup and restore capabilities, and security throughout the IT service lifecycle.

Datacenter Care for Multivendor Datacenter Care for Cloud Datacenter Care—Infrastructure Automation

||||||||||||||||||||

Activity: Proactive Care Advanced video Instructions: Watch the Proactive Care Advanced https://www.hpe.com/uk/en/services/proactive-care-services.html.

video

at:

Answer the following questions: –

What does the Account Support Manager provide?

–

What is enhanced critical incident management?

–

What hardware is included in the support?

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check Which resources can Foundation Care, Proactive Care, and Proactive Care Advanced customers access through the HPE Support Center?

||||||||||||||||||||

||||||||||||||||||||

HPE Financial Services Seventy-nine percent of surveyed respondents want to see pay-per-use options in the future that bundle equipment, software, services, and maintenance. Fifty-six percent of respondents cited limited flexibility regarding capacity options as the biggest challenge with their organization’s pay-per-use. HPEFS help technology users in every HPE customer segment make the most of their IT investment. HPE financing and financial asset management tools simplify financial lifecycle management and provide affordable solutions for customers’ IT needs. Financial Services also provide a wide range of remarketing services to maximize the value of older equipment. With Financial Services, HPE can purchase an enterprise’s existing IT assets and lease them to the customer for an agreed term, offering an affordable financial migration path to a consolidated solution. At the end of the lease term, the customer returns leased equipment to HPE and migrates to a new solution. Trade-in credits and customized removal of existing HPE and third-party assets facilitate the move to the latest technology. Cost-effective global logistics and transportation service support colocation of distributed systems. Note For more information regarding Financial Services, open the hyperlink. https://www.hpe.com/h20195/v2/GetPDF.aspx/4AA1-2354ENUC.pdf

HPE Financial Services adding value for channel partners HPEFS provide IT and financial expertise to help partners help customers develop flexibly acquire and invest in IT solutions to better meet their IT and business goals while optimizing their resources. It helps partners sell more and earn more. It facilitates more flexible customer IT acquisition and partner solution with one HPE team. HPE FS, shown in 4-14, enables increase cash flow through faster payment to Channel Partners. Immediate revenue recognition: Enable Channel Partners to receive payment. When HPE FS is involved in sale, HPE FS can pay resellers and distributors directly, enabling resellers to preserve their credit line with their distributors. Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 4-14 HPE Financial Services adding value for channel partners

Our promise to our partners HPEFS values Channel Partners, as shown in Figure 4-15, and responds quickly and accurately. Proactively establish credit lines for eligible customers to help facilitate future business. Pay Partners within two business days after customer acceptance. Provide expertise and a collaborative experience to accelerate transactions. Help you retain customers—more than 75% of HPEFS customers are repeat customers. Maintain an Open Door for ongoing, active dialogue with our Partners.

Figure 4-15 The HPE Financial Services

Leasing

||||||||||||||||||||

||||||||||||||||||||

Leasing allows customers to pay for technology capacity as needed. It leverages a baseline payment over a fixed term with incremental cost based on average monthly usage rates. Most important benefits are shown in Figure 4-16. The HPE Leasing Program allows customers to: Replace damaged, destroyed, stolen, or lost HPE and third-party hardware. Stay in business in event of disaster. Reduce total cost of ownership and increase return on investment. Conserve capital by avoiding a large upfront payment. Obtain total solution financing including hardware, software, service, and soft costs. Meet changing needs with maximum flexibility (add-ons and upgrades). Manage financial assets and track online assets with effective IT investment management tools. Eliminate end-of-useful-life hassles including disposal costs.

Figure 4-16 HPE Leasing Program

HPE Renew program The HPE Renew program offers a comprehensive portfolio of remanufactured HPE products that are like new, at competitive prices. HPE warranty is assured because all Renew products undergo the same building and testing process as new products. More than 80% of the Renew product portfolio is less than 12 months old. Inventory comes from various sources: Demonstration and trial equipment Lease and rental returns Discontinued products Excessive inventory from factory Customer returns

Technet24

||||||||||||||||||||

||||||||||||||||||||

Trade-in and trade-up programs If customers are looking for refurbished or reconditioned technology, remanufactured products from HPE Renew provide a meaningful option because they perform well and have the same warranty as new products. Customer benefits, shown in Figure 4-17, include: Guaranteed HPE quality and reliability Products backed by original HPE warranty, service, and support options Exceptional price versus performance Access to older technology to help with end-of-life transitions Configurable on demand for value products (business-critical services and storage) Straightforward buying procedures using standard HPE processes

Figure 4-17 HPE Renew program Customers should consider Renew if they: Have a limited IT budget and need same-as-new warranty. Are competing against a broker. Need legacy systems. Note Prices vary by region. Customers can contact HPE Renew for the latest pricing and products. Open the hyperlink for more information. https://hpe-renew.hpe.com/us/en/index.html

||||||||||||||||||||

||||||||||||||||||||

Activity: Understanding the Financial Services sales steps Instructions: Access the Partner Portal for this activity at https://partner.hpe.com. Select Products → Financial Services. Then, below HPEFS Partner Support Resources, click Learn more. Download the “HPEFS Channel Partner handbook.” Write down the seven steps to the sales process with Financial Services.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE Consulting Services Success in today’s mobile, connected, interactive, and immediate world requires that technology helps businesses create growth. Success demands transformative IT and business strategies and a highly flexible and responsive technology infrastructure. Leading organizations use technology to respond to business requirements and market opportunities quickly by overcoming aging, inflexible infrastructure. Specific HPE services can help customers realize their storage goals, including HPE enterprise, consulting, and support services. Support services include Foundation Care, Proactive Care, Proactive Care Advanced, and Datacenter Care.

Simplify IT with HPE Pointnext Services HPE Pointnext services make it easy to accelerate digital transformation. HPE Pointnext uses 25,000 professionals around the globe to help customers design, deploy, and optimize hybrid IT infrastructure. These professionals collaborate on around 11,000 projects per year, and the experience they have gained, coupled with an extensive library of infrastructure blueprints and design tools, helps to reduce risk for customers undergoing transformation. Working with a large cross-section of businesses and a rich ecosystem of partners, Pointnext professionals can help with advisory, professional, and operational services to find, implement, and support the optimal solution for customers’ business needs. HPE Pointnext Services include: Platform Consulting Services—HPE provides assistance in developing IT strategies to modernize and migrate to an on-premises automated, digital platform. Data Center Facilities Consulting Services—HPE helps in determining the best data center strategy to provide the right-sized requirements and providing the right mix of hybrid IT. Cloud Services—HPE provides an extensive catalog of cloud services to help customers define their strategy and roadmap for the journey to the cloud. Storage Services—HPE Pointnext storage consultants analyze and design storage systems to improve efficiency, reduce costs, and apply scalability to meet business needs, while keeping it all safe. Network Services—HPE helps businesses plan and manage the lifecycle of services.

||||||||||||||||||||

||||||||||||||||||||

This could be in deploying complex, business-critical solutions and helping to mitigate costly installation and configuration errors. Onsite services provide an end-to-end program that includes project management, implementation, and multivendor integration.

HPE Services portfolio Digital transformation requires continuous innovation, adoption, and evolution. HPE Pointnext, shown in Figure 4-18,can help you simplify the experience. We have thousands of global IT operational services and experts who will work with you on a daily basis. We will help you create new experiences for the business, from the core to the edge, offering new ways of delivering IT support services by automating and optimizing workloads, resources, and capacity, on-premises and in the cloud.

Figure 4-18 HPE Services portfolio

Professional Services Professional Services portfolio is shown in Figure 4-19. Access the link to show some more details around the different offerings. Select Storage Services and select the “Empower the data driven organization” brochure.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 4-19 Professional Services https://www.hpe.com/us/en/services/professional.html#servicesPortfolio

Operational Support Services Figure 4-20 shows the Operational Support Services portfolio. With Flexible Capacity, you get one contract and one monthly invoice based on consumption and usage with no upfront payment.

||||||||||||||||||||

||||||||||||||||||||

Figure 4-20 Operational Support Services Datacenter Care is a support solution that can cover your whole data center under one agreement, and you have only one number to call if there is a fault. HPE enables you to bridge the gap with its Edge Support IoT (Internet of Things) services portfolio. We help you set up and run your IoT efficiently and securely. Access the link to learn https://www.hpe.com/us/en/services/operational.html#servicesPortfolio

more:

Data Storage Services You need to change, but digital transformation is not simple. We have the expertise to help you do it. Our heritage in infrastructure, partner ecosystems, the end-to-end lifecycle experience, and our powerful, scalable IT solutions from Hybrid IT to the intelligent edge will help make change easy for you. The Storage consulting service portfolio, shown in Figure 4-21, offers a number of supporting service for customers who might wish to: Modernize their storage infrastructure. Perform a data migration for technical refresh. Continuity and Protection services to maintain the highest levels of High Availability.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 4-21 Data Storage Services Leverage the experience of an industry leader in digital era IT advisory and transformation services with HPE Pointnext. Our IT blueprints focus on your business outcomes and goals and consider your specific transformation needs. We listen carefully and will roadmap your strategy and design your transformation—tuned to your unique challenges—to help you drive better digital experience for you and for your customers.

HPE Solution Consulting Services

Figure 4-22 HPE Solution Consulting Services HPE Solution Consulting Services Portfolio includes: Platform Consulting Services Data Center Facilities Consulting Services Cloud Services Storage Services Network Services

||||||||||||||||||||

||||||||||||||||||||

Note More information is available https://www.hpe.com/us/en/services/consulting.html#servicesPortfolio

at

HPE 3PAR StoreServ Data Migration Service Available in convenient packaged offerings, the HPE 3PAR StoreServ Data Migration Service, shown in Figure 4-23, provides a flexible means of migrating organization’s critical data safely—with minimal impact to operation. Leverage HPE’s depth of technical capabilities, and industry experience, combined with proven migration methodologies, to help increasing OPEX (operating expense) and CAPEX (capital expenditure) savings and accelerate transition to HPE 3PAR StoreServ Storage. Note More information about the service is available in the data sheet: https://h20195.www2.hpe.com/V2/GetPDF.aspx/4AA4-6265ENW.pdf

Figure 4-23 HPE 3PAR StoreServ Data Migration Service

HPE Network Transformation Experience Workshop Technet24

||||||||||||||||||||

||||||||||||||||||||

Led by experienced HPE technology consultants, HPE Network Transformation Experience Workshop Service, shown in Figure 4-24, provides a highly interactive, meaningful day-long session (eight hours) with a customer’s IT, business, and executive stakeholders. Using a series of high-quality (slide-free) discussion panels, HPE will facilitate an exploration of a network transformation journey to business-aligned connectivity, aligning specific situation and HPE’s experiences. The HPE technology consultant will also lead discussions on the potential implications that network technologies such as software-defined networking (SDN), IPv6, and security might present to IT, network staff, and business. Note More information about the service is available in the data sheet: https://h20195.www2.hpe.com/v2/GetPDF.aspx/4AA5-4870ENN.pdf

Figure 4-24 HPE Network Transformation Experience Workshop

||||||||||||||||||||

||||||||||||||||||||

Learning check Write a summary of the key concepts presented in this chapter.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Summary The HPE Partner Ready Delivery Services program supports partners who sell or deliver HPE Services. The program includes designing and deploying new storage solutions, managing customer storage environments, and providing responsive support. Comprehensive support services from HPE include Datacenter Care, Proactive Care Services, Foundation Care, and Financial Services. HPE Solution Consulting Services span several areas, including analytics, data management, applications, data centers, workloads, and cloud. HPEFS help technology users in every customer segment make the most of their IT investment. HPE Pointnext provides services and consulting for the digital transformation age.

||||||||||||||||||||

||||||||||||||||||||

5

Planning HPE Storage Solutions

Technet24

||||||||||||||||||||

||||||||||||||||||||

OBJECTIVES After completing this chapter, you should be able to: Choose the most appropriate drive type for a customer’s needs. Describe planning a Redundant Array of Inexpensive Disk (RAID) configuration. Identify drive IO and performance. Describe storage area network (SAN) design principles and considerations. Explain planning a Hewlett Packard Enterprise (HPE) Flex-10 and FlexFabric design. Discuss considerations when configuring an HPE 3PAR StoreServ 9000. Provide an overview of HPE Reference Architectures.

||||||||||||||||||||

||||||||||||||||||||

Prelearning check When discussing concerns around life expectancy of solid-state drives (SSDs), what reassurance can you offer regarding HPE SSDs?

Technet24

||||||||||||||||||||

||||||||||||||||||||

Choosing the most appropriate drive type Enterprise-class storage must meet the maximum reliability and scalable performance goals of the 24/7 enterprise environment. Every minute of downtime can be costly, and slow performance can increase operating costs. Hard disks, controllers, interfaces, and drives have evolved to address the reliability and performance issues of the enterprise storage system. Some enterprise storage systems can be configured as tiered storage. Primary storage holds data that is in active use and might consist of hard disk drives (HDDs) or flashbased SSDs on a centralized SAN. You can also choose to create network-attached storage (NAS) or direct-attached storage (DAS). Planning and designing an enterprise storage system that can handle large volumes of data and a large number of users is a multistep process with many decision points along the way. Whichever architecture you choose, one of the first steps is to decide which type of drive to use in the storage arrays. Key considerations when choosing a drive type include: Performance Access speed Reliability Cost Drives are orderable at the time the array is purchased or can be added in the future when additional capacity is required.

Hard disk drives HDDs with rapidly spinning platters are the storage devices that have traditionally been used in personal computers and in the data center. The platters spin under a read/write head that moves on an actuator arm. Files are written and accessed randomly across the platters, so the head moves to various locations as the data is stored. Because HDDs, shown in Figure 5-1, use moving parts, they are slow and prone to mechanical failure.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-1 Hard disk drives

IO workloads There are four basic workloads, shown in Figure 5-2, that a storage array must be able to service: Sequential writes Random writes Sequential reads Random reads

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 5-2 IO workloads The first three workloads benefit from cache on a storage array controller because the algorithms designed into the array software leverage the array cache to service these workloads, improving IO for these workloads and hence reducing the IO latency experienced by a host. For write IOs (both sequential and random), data is mirrored in cache on two separate nodes in the array and then the server is told the IO is complete. This removes the need to flush the data to relatively slow drives before sending IO complete to the server. Caching of write data also allows for the grouping of small independent random writes into larger single IOs or even into full stripe writes, which reduces the total number of write IOs occurring to the back end of the array. In the case of sequential write workloads, not only are the writes acknowledged to the server after being mirrored across two nodes, but the array software can perform full stripe writes of the data to the HDDs on the back end of the array, making cache flushes very efficient. In the case of sequential read workloads, the array software will detect the sequential read nature of the IO stream and will begin to prefetch large amounts of data from HDDs on the back end of the array into read cache in anticipation of the server requesting the data in the near future and then being able to service these requests from read cache. The fourth workload, random reads, generally does not benefit from array cache and that is why these IOs are termed “random.” Because the data being read is random in nature, the array algorithms cannot anticipate the data that will be requested and have that data prestaged into read cache before a host requests it. As a result, random reads result in a lot of what are termed “read miss” or “cache miss” events where the requested data is not in read cache and must be retrieved from the back end of the array. So to satisfy a random read request, data has to come from the HDDs on the back end of the array, and this limits the number of random read input/output operations per second (IOPS) the array can deliver and also impacts the response time latency of those IOPS.

||||||||||||||||||||

Considerations while choosing HDDs There are many factors to consider while choosing HDDs: Rotational speed—Rotation speed is measured in RPM. Typical disk speeds for most enterprise workloads range from 10,000 RPM to 15,000 RPM. Seek time—Also known as access time, it measures the amount of time required for the read/write head to travel from its current location to the location needed to perform a read/write operation. The lower the seek time, the better the performance. Drive bus—The drive bus connects a drive to the disk controller. Examples include serial-attached SCSI (SAS), serial AT attachment (SATA), and Fibre Channel. IOPS—Input/output operations per second is the most commonly used metric for measuring a storage system’s overall performance. Several factors are used to calculate the IOPS of a storage system: –

Random performance—Random operations access non-adjoining locations on the hard drive and are generally associated with small data transfer sizes.

–

Sequential performance—Sequential operations access adjoining locations on the hard drive and are generally associated with large data transfer sizes.

–

A combination of the two, which is measured when you assess random and sequential performance at the same time.

||||||||||||||||||||

Write throttling—Write throttling is invoked when the drive is wearing at a rate where it would not last its warrantied five years. If a drive has worn more than expected, the write bandwidth is reduced. If the drive has worn less, more bandwidth is allowed. Cache—The small amount of cache memory embedded in the drive acts as a buffer to store the most recently accessed data. If the operating system requests data that is in the cache, the drive sends it immediately without waiting for an actual read operation. Cache is typically used for reads rather than writes, because during a write operation, data could be reported as written but lost during a power outage. The larger the cache, the better overall performance.

HDD types The two most common HDD types are SAS and SATA. SAS drives use a serial protocol to move data rather than a parallel bus technology. SAS drives use SCSI commands to

Technet24

||||||||||||||||||||

||||||||||||||||||||

perform error-recovery and error-reporting functions, and they are more effective than the commands used by SATA drives. HDD controllers use on-board data buffering and caching techniques to optimize the use of slow drives. Queuing techniques such as Tagged Command Queuing (TCQ) and Native Command Queuing (NCQ) allow controllers and compatible HDDs to take advantage of the position of the read/write head for more efficient drive operations. SATA drives use NCQ, which minimizes the distance the read head travels across the platters. SAS drives are available in speeds up to 15,000 RPM; SATA drives are available in speeds up to 10,000 RPM. SAS drives usually have better seek times than SATA drives, but SATA drives ship in larger capacities than SAS drives. Compared to 3.5-inch large form factor (LFF) SAS drives, 2.5-inch small form factor (SFF) SAS drives have these advantages: Lower annual failure rate (AFR) 30% smaller physical size Up to half the power consumption Increasing HDD density in a RAID-based system invariably enhances the overall performance of the storage system. Increasing the drive density also expands the choice of RAID strategies. SATA is a low-cost interface designed for point-to-point connection either through a cable or printed circuit board (PCB) trace. The host connection is to an advanced host controller interface (AHCI), which usually resides in the host chipset as the host adapter on the PCIe (PCI-Express, Peripheral Component Interconnect Express) bus. Architecture is shown in Figure 5-3.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-3 SATA communication (source: Seagate Technology, 2011) SAS is also a serial interface, attached to the host through a host adapter, but there are significant differences that make it suitable as an SSD interface: Less hardware overhead Faster transfer rates Wide ports Efficient driver-controller interfaces In addition, SAS includes features not found in SATA that improve reliability and availability of devices connected to the interface: Robust serial protocol Multiple host support End-to-end data integrity

Solid-state drives HPE SSDs are suited to enterprise environments with highly random data under a variety of write-workload applications. SSDs provide significantly better random read and write IOPS compared to HDDs. Although sequential read and write throughput is also improved over HDDs, the greatest benefit is recognized in random data applications. As a result, these high-performance, low-latency, and low-power SSDs provide significant system benefits for applications that previously overprovisioned HDD capacity to achieve better performance. SSDs use flash memory technology. They have no moving parts and do not experience the latency or synchronization issues that are drawbacks of HDDs. SSD-sustained data transfer rates are much faster than those of HDDs. They also consume much less power. HPE SSDs are available in categories based on their typical target workloads: Read intensive—HPE read-intensive 12G SAS and 6G SATA SSDs deliver enterprise features for a low price in HPE ProLiant server systems. This entry-level pricing is fueling rapid SSD adoption for read-intensive workloads because the cost per IOPS compares very favorably to HDDs. Read-intensive SSDs deliver great performance for workloads high in reads such as boot/swap, web servers, and read caching. Mixed use—HPE mixed-use 12G SAS and 6G SATA SSDs are best suited for high IO applications with workloads balanced between reads and writes. The SAS and SATA Technet24

||||||||||||||||||||

||||||||||||||||||||

SSDs provide the workload-optimized performance required for demanding IOintensive applications. When paired with ProLiant servers, these SSDs help you meet the challenges of big data. They achieve twice the performance and endurance of previous HPE SAS and SATA SSDs. The SATA SSDs ship with a 6Gb/s SATA hot-plug interface. Write intensive—HPE write-intensive 12G SAS SSDs provide high write performance and endurance. They are best suited for mission-critical enterprise environments with workloads high in both reads and writes. Workloads best suited for these SSDs include online transaction processing (OLTP), virtual desktop infrastructure, business intelligence, and big data analytics. These categories include both SAS SSDs and SATA SSDs. The categories indicate the number of drive writes per day (DWPD) that you can expect from the drive. Note DWPD is the maximum number of 4K host writes to the entire drive capacity of the SSD per day over a five-year period. Read-intensive SSDs are typically available at the lowest price with an endurance of less than or equal to 1 DWPD. Write-intensive SSDs typically have the highest write performance, with a typical endurance of 10 DWPD or more. Mixed-use SSDs are for workloads that need a balance of strong read and write performance, with endurance typically more than 1 and less than 10 DWPD.

Choosing an HDD or SSD solution for an application To meet the demands for virtualized and data-intensive workloads, HPE optimized the storage architecture of the latest generation ProLiant servers for SSD performance, energy efficiency, and high density. The criteria for choosing SSD solutions are very strict. For example, consider how well NAND flash manages bad blocks of memory, wear leveling, and write amplification. These characteristics greatly affect SSD endurance. The HPE 3G SATA SSDs are available in various capacities for ProLiant servers. HPE 100 GB SSD pricing provides a competitive cost alternative with 146 GB 15,000 RPM HDDs and other lower capacity HDDs that are approaching the end of their life span. Matching storage solutions with application-environment requirements goes beyond

||||||||||||||||||||

||||||||||||||||||||

looking at specifications. It requires knowledge of current and future business needs. Comparison of HDD and SSD is shown in Figure 5-4.

Figure 5-4 Solid-state drive technology—HDD vs. SSD The HPE 3PAR 7-year support covers all deployment types, including all-flash and hybrid models, as shown in Figure 5-5.

Figure 5-5 Solid-state drive support

Comparing drive performance SSDs provide significant performance advantages over traditional HDDs. Figure 5-6 shows the relative performance in IOPS and latency for a performance HDD, a mainstream SSD, and an IO accelerator. The HPE IO Accelerator is an interface card containing an SSD, flash memory, and a custom controller that significantly improves Technet24

||||||||||||||||||||

||||||||||||||||||||

data throughput with external storage devices.

Figure 5-6 Comparing drive performance SSDs do not have seek or rotational latency time. They address any sector of the NAND flash directly in 0.1 millisecond. SSD latency includes the time for memory access and transfer combined with controller overhead. SSDs excel at random read operations, where their performance can be more than 100 times better than that of spinning media drives. SSDs perform random writes at least 25 times faster than a comparable 15,000 RPM HDD. This means SSDs provide improved application performance. In Figure 5-6, for each vertical bar, the top number indicates IOPS for 100% random reads and the bottom number indicates IOPS for 100% random writes. It shows how the mainstream SSD and IO Accelerator vastly outperform the HDD and are capable of reading and writing data much faster. Figure 5-7 compares the performance of different storage devices.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-7 High performance SSD vs. HDD technology

NAND flash memory types NAND flash memory is a type of nonvolatile memory that retains data even when power is turned off. The development of NAND flash memory has reduced the cost per bit and increased maximum chip capacity. As a result, flash memory is competitive with HDDs in terms of cost, performance, and reliability and is replacing HDDs in many environments. NAND flash memory is written and read in 128 KB blocks, which are the smallest erasable units in a NAND device. Several types of NAND flash memory are available: Single-level cell (SLC)—Stores one bit per cell and is in general fast, reliable, and expensive. Multilevel cell (MLC)—Offers greater storage density than SLC memory but is in general slower and less robust. Enterprise multilevel cell (eMLC)—Stores two bits per cell, has greater endurance than MLC memory, and uses an advanced controller. Consumer-grade multilevel cell (cMLC)—Stores two bits per cell; is the lowest-cost NAND flash memory type; and is commonly used in cameras, cell phones, and USB drives. Three-level cell (TLC)—Stores three bits per cell and has more power and better error correction than other types of NAND flash memory. The newest NAND technology is 3D NAND, in which memory cells are stacked vertically in multiple planes in the chip. This architecture increases the number of bits per chip to offer more storage capacity in a smaller physical space. It also reduces Technet24

||||||||||||||||||||

||||||||||||||||||||

power consumption because the interconnect length between cells is shorter.

Wear protection technology NAND flash devices use semiconductor technology that supports a finite number of data writes to the device, defined as the maximum lifetime. Wear-leveling algorithms maximize the endurance or life span of SSDs. This technology remaps logical memory blocks receiving frequent writes to different physical pages. It evenly distributes erasures and rewrites across the storage medium. A pointer array on the SSD controller contains the logical-to-physical map. ProLiant servers ship with HPE SmartSSD Wear Gauge to maximize SSD media utilization and eliminate unplanned downtime. The SmartSSD Wear Gauge utility monitors and reports the percentage of a drive’s lifecycle used and the amount of life remaining under the workload to date. When the SSD reaches its maximum rated usage limit, notification is sent to operating system event logs and managements agents. That way, the SSD can be replaced before it fails and reduces the risk of data loss. To view the SmartSSD Wear SmartSSD, run the Smart Storage Administrator (SSA).

Data deduplication on HPE 3PAR StoreServ SSDs With the increasing use of flash, deduplication for primary storage arrays has become essential. The cost differential between SSDs and HDDs requires compaction technologies such as thin provisioning and deduplication to make flash-based media more cost efficient. Widespread deployment of server virtualization is also driving the demand for primary storage deduplication. The potential benefits of deduplication correlate directly with data redundancy. VM (virtual machine) images and client virtualization environments with hosted virtual desktops are both characterized by a high degree of data redundancy, meaning that these are two use cases where primary deduplication fits well. Particularly in virtualized environments, primary storage arrays are subjected to unpredictable performance demands that can require simultaneously low latency and high throughput. This issue can be resolved with deduplication. The impact of deduplication on performance is determined by various parameters such as whether deduplication takes place as an in-line or background process and the level of granularity used for operations. Therefore, duplicating data at a fine level of granularity and simultaneously delivering space savings improvements generally requires a lot of CPU processing power and memory—more than most primary storage arrays have to spare. This can force trade-offs that restrict the overall efficiency of primary deduplication.

||||||||||||||||||||

||||||||||||||||||||

Enterprise data centers running a mix of legacy, high-end, all-flash arrays with limited scale have been forced to combine these silos with added layers of cost, risk, and complexity in the form of dedicated external SAN virtualization appliances placed directly in the data path. To eliminate these compromises, HPE announced enhancements to the native storage federation capabilities within HPE 3PAR StoreServ systems. Customers can achieve one-click workload rebalancing across arrays with no added overhead.

Drive IO and RAID performance impact The IO performance of storage media is affected by a number of factors including the RAID configuration. The random or sequential nature of the IO pattern significantly effects the amount of IO the drive can provide this is also affected by the block size and whether the IO is read or write. This is true for both Flash (SSD) and Spinning (HDD) media and must be considered when sizing an array. The capability of the interface (SAS, SATA, and PCIe) will affect the speed and feeds. SSD’s and Accelerator products are based on Nand Flash technology where IO performance is considered high compared to that of traditional mechanical spinning (HDD) media: Nearline disks are typically the slowest at 7.2k RPM; these disks will yield around 75 IO per second per spindle. Fast Class disks on the other hand come in 10k and 15k RPM and yield a higher 150 and 200 IO per second per spindle, respectively. However, solid-state disks and accelerators can yield 10s of thousands of IO per second for certain IO patterns. Figure 5-8 shows performance value for an example drive, HPE 960GB SATA Mixed Use SSD.

Figure 5-8 Drive IO and RAID performance impact

Technet24

||||||||||||||||||||

||||||||||||||||||||

Today, SSD disks are classified into 3 different workload types: RI—read intensive, typically Databases, Web Servers, Email, Analytics, <= 1 write per day WI—write intensive, typically OLTP, Data Warehousing, ERP, database logs, >1 <10 writes per day MU—mixed use, typically Balanced resources for Business Intelligence and Transactions, >10 write per day RAID performance for read IO is about the same across all RAID levels but write penalties apply: RAID 1, as a new write comes in, data and mirror are written out (two write IOPS). RAID 5, as a write comes in, the old parity and old data must be read (two read IOPS), and then parity calculated and new parity and new data written out (two write IOPS). Therefore, each random write to R5 has four backend IOPS (two read and two write). RAID 6, as a write comes in, both old parities and the old data must be read (three read IOPS), and then parity is recalculated, and two new parities and new data are written out (three write IOPS) Therefore, each random write to R6 has six backend IOPS (three read and three write). Large sequential writes can write full RAID stripes for R5 and R6 without the need for recalculations. While RAID 6 has the biggest impact on performance in heavy write intensive environment, it also offers the highest level of redundancy; this is why it is the recommended RAID for spinning (HDD) disks. SSDs and HDDs use SAS or SATA protocols to interface with the host system, but SSDs store and retrieve data in flash memory arrays rather than on spinning media. You can use SSDs with HPE Smart Array controllers in redundant RAID configurations. In RAID 5 and RAID 6 configurations, SSDs have several unique advantages. Both RAID 5 and RAID 6 require the Smart Array controller to perform two read-modifywrite operations to the backend array drives for every host-level random write it executes. With HDDs, the performance penalty for using RAID 5 and RAID 6 is severe. Each read-modify-write operation requires an extra revolution of the disk media, causing the system to incur the worst possible latency. For these reasons, RAID 1 or RAID 10 is often configured with HDD arrays. These RAID levels achieve redundancy without suffering as a great performance penalty. With SSDs, however, there is no rotational latency. As a result, you can create RAID 5 or RAID 6 arrays using SSDs that have about the same performance as RAID 1 or RAID

||||||||||||||||||||

||||||||||||||||||||

10 arrays.

Adaptive Flash Cache HPE 3PAR Adaptive Flash Cache (AFC), shown in Figure 5-9, is included as part of the HPE 3PAR OS Suite 3.2.1 and later and is supported on all HPE 3PAR StoreServ arrays that have a mix of both SSDs and HDDs.

Figure 5-9 Adaptive Flash Cache AFC leverages SSD capacities as read cache extension for spinning media and can be enabled system wide or for individual virtual volume sets (VVsets). If customers require a lot of small read blocks, they could benefit from AFC because: AFC enhances HPE 3PAR Adaptive Optimization.

Technet24

||||||||||||||||||||

||||||||||||||||||||

AFC increases read cache hits by destaging 16 kB dynamic random access memory (DRAM) pages to flash cache in real time instead of purging. AFC lowers latency for random read-intensive IO workloads. AFC has faster response times for read bursts on HDD and cold data on AO volumes. AFC does not require dedicated SSDs; they can be shared with the SSD tier and AO. Minimum SSDs required for AFC only use two SSDs per node pair. Maximum AFC capacity, 32 TiB (for 20000 array). No license is required; AFC is included in the HPE 3PAR OS software. The HPE 3PAR Management Console built-in simulator for AFC helps determine the expected acceleration. AFC provides read cache extension by leveraging HPE 3PAR virtualization technologies. This functionality allows dedicating a portion of SSD capacity as an augmentation of the HPE 3PAR array primary cache, reducing application response time for read-intensive IO workloads. It also improves performance when periodically reading data from a slower tier of HDDs. AFC works by caching the “hot” data: Data is fetched into DRAM from HDDs to fulfill host requests. DRAM data is copied (admitted) into flash cache when removed from DRAM. Future host reads are redirected to and served from flash cache. Data in flash cache is assigned a different “temperature”; hot data stays and cold data is removed (evicted) in the background. HPE 3PAR Autonomic Rebalance provided via the tunesys operation does not apply to Flash Cache logical drives. After adding additional SSDs to a system, Flash Cache can be rebalanced across the new SSDs by first removing Flash Cache with the removeflashcache command (This will disable Flash Cache for all VVs.) and then recreating Flash Cache with the createflashcache command.

Using AFC to increase read hits and improve random read performance If an array’s read cache is sufficiently large compared to its working set size, it is possible to turn a percentage of what would have been random read misses or cache

||||||||||||||||||||

||||||||||||||||||||

misses into read hits where the data being requested by a server is already in read cache. One of the easiest ways to achieve this is to add more DRAM to an array to increase the size of read cache, but this is an expensive option. A better solution is to use SSDs as a Layer 2 (L2) read cache to hold small-block random-read data that is being removed from DRAM cache with the expectation that some percentage of that data will be reread. This is a more cost-effective way of keeping a larger percentage of read data on very fast media to improve overall randomread performance. This is exactly what HPE 3PAR Adaptive Flash Cache does; it uses flash storage (SSDs) as L2 cache to hold random read data that is being removed from DRAM cache, keeping it on high-performance flash media longer in the event the data is accessed again by a host.

What gets cached in AFC? With Adaptive Flash Cache, only small-block random-read data (IO size less than 64 KiB) that is resident in a node’s DRAM cache can be considered for caching in the AFC. Data is not cached directly from spinning media into AFC; it only gets placed into AFC after having been resident as read data in a node’s DRAM cache first. Data is prefetched into DRAM cache as a result of a sequential read-ahead stream (regardless of the IO size). Data that ends up in DRAM cache (for IOs that are 64 KiB or larger) is not considered for caching into AFC. Although AFC is not used as an extension of write cache for write data, it is possible for data that ends up in DRAM memory as a result of a small-block random-write operations (IO size < 64 KiB) to be destaged to AFC. This occurs after dirty write data is written to disk, as during a cache flush operation, and the Cache Memory Page (CMP, 16 KiB page of physical DRAM cache memory that contains data for either a read IO or a write IO) becomes marked as a valid page of read data in the DRAM cache (a CMP containing clean, valid data). At this point, the CMP containing the data is treated by the array caching algorithms as if it were a page of small-block random-read data, and it will ultimately find itself in AFC when it is eventually removed from DRAM cache. Sequential write data will not be destaged to AFC regardless of the IO size, and random write data for IOs that are 64 KiB or larger will not be destaged to AFC even though write data for a sequential write IO stream also becomes read data when flushed to the back end of the array. Data that is read into DRAM cache from an SSD tier will never be cached in AFC because that data is already on a high-performance flash tier and would not benefit from being in AFC.

Data in AFC

Technet24

||||||||||||||||||||

||||||||||||||||||||

When a page of read data moves from DRAM cache to AFC, it is copied to a Flash Memory Page (FMP) that resides on one of five least recently used (LRU) queues that are maintained by AFC. These LRU queues are used for tracking how hot (frequently accessed) data is as it resides in AFC. FMPs move from lower priority LRU queues to higher priority queues when they are accessed (read) by a host, and they move from higher to lower LRU queues when LRU queues are demoted. If an FMP continues to be accessed once it gets into AFC, it will continually be promoted to a higher priority LRU and can, in fact, stay in AFC indefinitely. There are five different LRU queues maintained by AFC: Dormant—This is the lowest priority LRU queue and holds FMPs that contain invalid data or valid data that is a candidate for reuse because of lack of access since entering AFC. These FMPs are used when a new DRAM CMP, which does not already have an entry in AFC, needs to be copied to AFC. A dormant LRU FMP is immediately promoted to the norm LRU queue when it is used to hold a new page of data entering AFC. Existing valid FMPs in this LRU are promoted to warm if they are accessed by a server. Cold—FMPs are never promoted from dormant to cold. FMPs in this LRU queue are promoted to warm if they are accessed. FMPs are moved here from norm when all LRU queues are demoted down. Norm—When a DRAM CMP is initially destaged to AFC, it starts in this LRU queue. FMPs in this LRU queue are promoted to warm if they are accessed once. Warm—FMPs from this LRU queue are promoted to hot if they are accessed twice. Hot—FMPs from the warm queue are promoted to hot if they are accessed twice and they stay in the AFC on subsequent hits. When a server reads data in AFC, it is copied to DRAM read cache where it will, again, eventually be removed. However, it does not have to be copied to an LRU queue when it is removed. The data already resides in AFC on the LRU queue it was promoted to because of the host read. If data sits in AFC long enough without being accessed, it will eventually be aged out because of LRU demotion or be reused for a new DRAM CMP entering AFC.

Moving data from a node’s DRAM read cache into AFC Data in a node’s DRAM read cache starts to destage to AFC when the node’s DRAM memory is 90% full. Read data is copied from a node’s DRAM cache to AFC when the caching algorithms are looking to free up CMPs in a node’s DRAM memory and a read

||||||||||||||||||||

||||||||||||||||||||

CMP is selected to be freed up, as shown in Figure 5-10.

Figure 5-10 Moving data from a node´s DRAM read cache into AFC

HPE SmartCache architecture Storage needs for enterprise systems are advancing rapidly in the areas of performance and capacity, and most applications require a balance of these two attributes. However, certain applications emphasize one attribute over another. For instance, applications that process a lot of read and write requests improve performance. For applications that manage continually growing archival data, capacity is vital. Two storage technologies commonly address application needs. The SSD, with its very low latency, is ideal for performance-conscious applications. The HDD, with its longterm reliability and economical cost per GB, meets large capacity needs. This figure shows the suitability of SSDs and HDDs for meeting the needs of different applications. A few important notes about this figure include: High-frequency trading applications require lower latency. Technet24

||||||||||||||||||||

||||||||||||||||||||

An all-SSD solution would provide the performance this environment requires. OLTP and analytics push the boundaries of both performance and capacity growth. A mix of SSDs and HDDs best meets those needs. Archival and smartphone applications require unstructured data growth, a need that can be addressed with high-capacity HDDs. HPE SmartCache is a controller-based caching solution in a DAS environment that caches the most frequently accessed data ("hot" data) onto lower latency highperforming SSDs to dynamically accelerate application workloads. SmartCache can be enabled on specific HPE Smart Array P-Series controllers found in HPE ProLiant Gen8 and above servers. With SmartCache, HPE estimates that customers can gain up to six times better performance with transactional workloads and 50% better performance for video-streaming applications. The collaborative-caching solution with SmartCache and HPE 3PAR autonomically copies data in real time from the HPE 3PAR array to the HPE Smart Array SSD cache on ProLiant Gen8 and above servers. Autonomic is a word used to describe when a system can automatically adjust and adapt to its environment. SmartCache combines different technologies and device types to close the cost/performance gap. SmartCache uses a caching architecture where a copy of the data resides on the HDD and on a lower latency device that is used for caching. The basic HPE SmartCache architecture, shown in Figure 5-11, includes these three elements: Bulk storage—The bulk storage device can be either HDDs or connections to SAN storage. Accelerator—The accelerator is a faster/lower latency device that caches data. Its capacity is less than the capacity of the bulk storage device. Metadata—The metadata is information held in a relatively small storage area that maps the location of information residing on the accelerator and bulk storage devices.

Figure 5-11 HPE SmartCache architecture The SmartCache architecture is flexible and supports numerous device types for bulk storage, accelerator, and metadata. This architecture can accommodate the evolution of

||||||||||||||||||||

||||||||||||||||||||

storage devices. SmartCache solutions are designed for DAS and SAN systems.

SmartCache—DAS SmartCache is hosted on HPE Smart Array controllers. It does not require a drive or application changes (for example, changes to the database). The metadata is stored within a portion of flash backed write cache (FBWC). The SmartCache DAS solution includes the three elements of the SmartCache architecture, as shown in Figure 5-12: HDDs serving as bulk storage SSDs as the accelerator FBWC memory for storing metadata

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 5-12 SmartCache—DAS The SmartCache control layer resides in the firmware of the onboard Smart Array controller of the ProLiant Gen8 server, below the operating system and driver. This allows caching for devices connected to a single array controller. SmartCache offers flexibility in creating logical disk volumes from HDDs: The accelerator or cache volume is designed to support any RAID configuration supported by the Smart Array controller. Each logical disk volume can have its own cache volume or none at all. Cache volumes can be created and assigned dynamically without adversely impacting applications running on the server. Note HPE SmartCache consists of firmware that provides the caching feature within the Smart Array controllers and requires a license key for activation. Only SSDs can be used for cache volumes, and a cache volume might be assigned only to a single logical disk volume. The SmartCache solution consumes a portion of the FBWC memory module on the Smart Array controller for metadata. To ensure sufficient storage for accelerator and metadata, you should recommend 1 or 2 GB of FBWC memory or 1 GB of metadata space for every TB of accelerator space. HPE recommends using either: 1 or 2 GB of FBWC memory 1 GB of metadata space for every terabyte of accelerator space When SmartCache is used for DAS, legacy cache is still present and operational. It uses the remaining space in the FBWC memory. When using SmartCache, you should set the legacy cache for 100% write operation. This allows write-back support on the Smart Array controller, which accelerates write performance. This also allows SSDs configured as SmartCache volumes to provide much larger read cache. Because SmartCache is implemented within the Smart Array firmware, it can be used on any operating system supported by ProLiant servers. This solution operates transparently, without any dependencies on operating system, device driver software, file system type, or applications. SmartCache supports write-through caching. When an application writes data to the disk, the Smart Array controller writes the data to the HDDs. If the data is also in the cache volume, the data is written to the SSD.

||||||||||||||||||||

The SmartCache solution supports write-through caching. When an application writes data to the disk, the Smart Array controller writes the data to the HDDs, and if the data is also in the cache volume, the data is written to the SSD.

Hybrid flash rules and guidelines for Nimble For good read latency target 8%–12% Flash to Usable Disk Capacity (FDR—Flash to Disk Ratio) Minimum Flash Disk Ratio is around 3%–4%. These configurations can be perfectly fine for DR target arrays. To enable deduplication on a Hybrid Flash system, you need a minimum of 8% FDR to contain the metadata (Available with NimbleOS 5.0 and above) 12/16% for dedupe. To hit the max sequential performance of CS7000, and the AF5000 and above, use a minimum of 4 active ports for array connectivity (6 active ports for AF9000). If a high level of sequential performance is required, consider higher performing platforms than AF1000/AF3000 and CS1000/CS3000 (5000 or above). Rules of Thumb –

Daily data change rate 1%–2% per day conservative estimate

–

Annual data growth rate 10%–30% per year

–

Data Reduction with compression only—ESX 1.5X, SQL 2X, Exchange 1.4X

–

Data Reduction with dedupe and compression—ESX 3X, SQL 2.8X, Exchange 1.75X

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check When discussing concerns around the life expectancy of SSDs, what reassurance can you offer regarding HPE SSDs?

Can frequently accessed data read from any disk tier to DRAM be moved into AFC?

||||||||||||||||||||

||||||||||||||||||||

Activity: Choosing the correct SSD Instructions: Browse to the HPE SSD Selector Tool at: http://ssd.hpe.com/ for this activity. Select the Help me choose option. From the middle column, select Database (mid-range), and click Next. Check Read Intensive, and click Next. Follow the screens to select a ProLiant DL380 Gen10 with 6 TB capacity using SFF SAS. Identify the chosen drive type, and view the QuickSpecs. Notes:

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Planning a RAID configuration The RAID configuration you choose for the storage system depends on the customer’s plans for data fault tolerance, data availability, and capacity growth. Note Plan your RAID configuration carefully for many traditional storage systems. After you have configured RAID, you cannot change the RAID configuration without deleting all data on the storage system. This does not apply to an HPE 3PAR StoreServ array. RAID protects against failure of disks within a storage system, but not against failure of an entire storage system. For example, if network connectivity to the storage system is lost, then data reads and writes to the storage system cannot continue. Keeping multiple copies of the data ensures that data is safe and remains available in the case of disk failure. There are two ways to achieve data protection: For StoreVirtual or StoreServ arrays, configure RAID 1, RAID 10, RAID 5, RAID 5 + Spare, RAID 50, or RAID 6 within each storage system to ensure data redundancy. In a StoreVirtual system, always use Network RAID to mirror data volumes across storage systems in a cluster, regardless of RAID level, for added data protection and high availability. Within each storage system, RAID 1 or RAID 10 ensures that two copies of all data exist. If one of the disks in a RAID pair goes down, data reads and writes continue on the other disk. Similarly, RAID 5, RAID 50, or RAID 6 provides redundancy by spreading parity evenly across the disks in the set. If one disk in a RAID 5 set or two disks in a RAID 6 set go down, data reads and writes continue on the remaining disks in the set. In RAID 50, up to one disk in each RAID 5 set can go down, and data reads and writes continue on the remaining disks. Note If you plan on using clusters with only a single storage system, use RAID 1 and RAID 10, or RAID 5. For spinning media, HPE recommends using RAID 6 to ensure the highest level of data redundancy within that storage system.

Technet24

||||||||||||||||||||

3PAR StoreServ RAID 6 concept RAID 6, shown in Figure 5-13, characteristics, features, and benefits include: RAID 6 uses double parity. Can rebuild data into sparing area (spare chunklets) in the event of double disk failure. A RAID 6 volume can survive the loss of up to two physical disks (data could be lost if three or more disks fail). RAID 6 is highly recommended for nearline (NL) and fast class (FC) disks and 15.35 TB SSDs. Beginning with HPE 3PAR OS 3.3.1, RAID 6 is the default raid level when creating FC and NL Common Provisioning Groups (CPGs). RAID 6 supports set sizes as follows: –

||||||||||||||||||||

6 (4D+2P), 8 (6D+2P), 10 (8D+2P), 12 (10D+2P), 16 (14D+2P)

RAID 6 incurs a 6 IO backend penalty on writes.

Figure 5-13 3PAR StoreServ RAID 6 concept On a RAID Multi-Parity (MP) or RAID 6 logical disk, data is striped across rows of RAID MP sets. A RAID MP set, or double-parity set, must contain at least 6 chunklets. A RAID MP set with 6 chunklets has a total of 4 chunklets of space for data and 2 chunklets of space for parity. Many RAID 6/MP set sizes are supported. The data and parity steps are striped across each chunklet in the set. The chunklets in each RAID MP set are distributed across different physical disks, which can be located in different drive magazines or even different drive cages. The number of sets in a row is the row size. The system accesses the data from a RAID MP logical disk in steps. The step size varies and is dependent on the size of the RAID MP set. A RAID MP set can function with the loss of any two of the chunklets in the set. For HPE Nimble Storage arrays, the RAID is preconfigured as triple parity, shown in Figure 5-14, and cannot be modified.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-14 AF-Series: Triple+ Parity with Integrated Spare

Using network RAID in a StoreVirtual Cluster A cluster is a group of storage systems across which data can be protected by using

Technet24

||||||||||||||||||||

||||||||||||||||||||

Network RAID. Network RAID protects against the failure of a RAID disk set within a storage system, failure of an entire storage system, or external failures such as networking or power. For example, if an entire storage system in a cluster becomes unavailable, data reads and writes continue because the missing data can be obtained from the other storage systems.

Using disk RAID with network RAID in a cluster Always use Network RAID in a cluster to protect volumes across storage systems. The redundancy provided by RAID 10, RAID 5, RAID 50, or RAID 6 ensures availability at the storage system level. Using Network RAID for volumes in a cluster ensures availability at the cluster level. For example: Using Network RAID, up to three copies of a volume can be created on a cluster of three storage systems. The Network RAID configuration ensures that two of the three storage systems can go offline and the volume is still accessible. Configuring RAID 10 on these storage systems means that each of the three copies of the volume is stored on two disks within the storage system, generating six copies of each volume. For a 50 GB volume, 300 GB of disk capacity is used. RAID 5 and RAID 50 use less disk capacity than RAID 1 or RAID 10, so they can be combined with Network RAID and still use capacity efficiently. One benefit of configuring RAID 5 or RAID 50 in storage systems that use Network RAID in a cluster is that if a single disk goes down, the data on that storage system can be rebuilt using RAID instead of requiring a complete copy from another storage system in the cluster. Rebuilding the disks within a single set is faster and creates less of a performance impact on applications accessing data compared to copying data from another storage system in the cluster. RAID 6 provides similar space benefits to RAID 5, with the additional protection of being able to survive the loss of up to two drives. If you are protecting volumes across a cluster, configuring the storage system for RAID 1 or RAID 10 consumes half the capacity of the storage system. Configuring the storage system for RAID 5 or RAID 50 provides redundancy within each storage system and allows most of the disk capacity to be used for data storage. RAID 6 provides greater redundancy on a single storage system but consumes more disk space than RAID 5.

||||||||||||||||||||

||||||||||||||||||||

Planning and designing a SAN A SAN provides the data communication infrastructure for advanced, cost-efficient storage systems. SAN technology offers investment protection, management features, and IO price performance to minimize capital expense. HPE SAN architecture provides open network storage solutions for all sizes and types of businesses, including small to medium-sized IT departments and enterprise environments. Fibre Channel currently supports a maximum of 239 switches in a single fabric. HPE specifies support based on rules for the maximum number of switches and ports in a single fabric or multifabric SAN. Using many switches to obtain a high number of ports is unacceptable if the fabric exceeds the total switch count limit. Similarly, using largecapacity switches can create a network that exceeds the maximum number of ports. SANs provide flexibility in system management, configuration, connectivity, and performance to meet the needs of the changing business environment. For the most challenging IT problems, SANs offer resilient solutions: Open systems—SANs support various operating systems and servers to meet operational requirements. A robust storage infrastructure accommodates new business models, unexpected growth, and corporate reorganizations. Fast backup and restore—SANs remove backup and recovery traffic from the LAN, reducing congestion, improving backup windows, and efficiently using storage resources. Centrally managed, high-performance tape libraries can be used to reduce backup overhead. Business continuance—SANs can eliminate single points of failure, incorporate failover software, and support mirroring at geographically dispersed data centers for disaster recovery. Customers can quickly restore productivity after a power failure or component downtime. High availability—Redundant fabric designs, storage replication, dynamic failover protection, traffic rerouting, and server clustering enable SANs to provide enterpriseclass availability to open system servers. Server and storage consolidation—Multiple servers and backup systems can share storage for efficient processing and increased availability. Centralized management—You can manage consolidated storage by using web-based tools from any location, thus reducing labor costs. Security—SANs

support network security measures

such as

authentication,

Technet24

||||||||||||||||||||

||||||||||||||||||||

authorization, access control, and zoning. Online scalability—Customers can add storage capacity or expand the fabric as needs change. Servers can be added or removed, and storage can be increased, changed, or reassigned when the SAN is online. Modularity—Modular design simplifies SAN scalability and increases return on investment by consolidating and sharing systems.

HPE approach to SAN design HPE has three approaches to SAN design, including: Standard HPE design—Specify the arrangement of Fibre Channel switches in a SAN fabric. HPE standard designs are optimized for specific data access requirements and typical workloads. Implementing a standard design is the simplest approach to SAN design. This approach is recommended for users who are designing a SAN for the first time. Modified HPE standard design—Select a standard SAN design that satisfies most requirements, and then modify it to meet data access and connectivity requirements. This approach is recommended for users with an intermediate level of SAN experience. Custom design using HPE SAN design rules—Use a custom SAN design for specific storage and data access requirements. HPE SAN design rules specify guidelines for configuring custom topologies. This approach is recommended for users with an intermediate or advanced level of SAN experience. When planning and preparing for the installation of a storage system, you assume the following responsibilities: Providing suitable space for unpacking, installing, and operating the storage system Maintaining the proper environmental conditions for the storage system Providing adequate power facilities for the storage system Supplying the network connections and external cabling required

SAN design principles and considerations To design or modify a SAN, evaluate these factors: Geographic layout—The locations of campuses, buildings, servers, and storage

||||||||||||||||||||

||||||||||||||||||||

systems determine the required SAN connections. SAN infrastructure components support long-distance connections and multiple interswitch cable segments. Fibre Channel routing connects independent SAN islands (fabrics) or virtual SANs (VSANs) to form a single, geographically distributed SAN. Data availability—A resilient SAN environment minimizes vulnerability to fabric or device failures and maximizes performance. A mixture of availability levels can be implemented in the same SAN, depending on the level of protection required for specific applications or data. Connectivity—Provide enough ports to connect servers, storage systems, and fabric components. To create a high-capacity SAN, you can connect multiple fabrics or VSANs using routing. Storage capacity—Calculate the total storage capacity requirement and determine the type and number of storage systems needed for current and future requirements. Note For detailed information, see the HPE SAN Design Reference Guide. To access the guide, open the hyperlink: http://h20564.www2.hpe.com/hpsc/doc/public/display? docId=c00403562&lang=en-us&cc=us

SAN components A SAN consists of several components, as shown in Figure 5-15.

Figure 5-15 SAN components A SAN consists of the following components: Switches—Fibre Channel switch creates the fabric of the SAN, including routers, bridges, and gateways.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Storage devices—This includes storage system types, such as disk arrays and tape libraries. Servers and HBAs— host bus adapters (HBAs) connect the server to the SAN. Transceivers, cabling, and cable connectors—Transceivers and fiber optic cables provide the physical connections between components. A SAN uses multiple paths to connect servers and storage systems. To take full advantage of its capabilities, the SAN is maintained separately from parallel generalpurpose networks. The network topology is the physical arrangement of interconnected hardware components. In a basic topology, a Fibre Channel switch interconnects multiple servers and a storage system. To protect against hardware failure, high-availability topologies connect redundant systems.

Topology data access performance To choose a SAN fabric topology, you must determine which data access type is appropriate for your environment. The data access types are as follows: Local (one-to-one)—Data access between a local server and a storage system connected to the same switch Centralized (many-to-one)—Data access between multiple, dispersed servers and one centrally located storage system Distributed (many-to-many)—Data access between multiple, dispersed servers and storage systems For mission-critical applications, HPE recommends that you implement a fully redundant fabric configuration. You can justify the additional cost if you consider the cost of losing access to critical data. With today’s large port count switches, just licensing the ports required allows SAN scaling with few switches.

Fibre Channel and converged switches End-to-end FCoE (Fibre Channel over Ethernet) is supported through FC ISLs (inter-

||||||||||||||||||||

||||||||||||||||||||

switch links). FCoE traffic can be carried across multiple FC ISL hops, eventually terminating in FC or FCoE storage. Support multiswitch networks with up to 30 switches and 7 hops (FCoE) or 3 hops (Fibre Channel). Convergence—Many HPE FlexFabric switches have 10GbE, 40GbE, or converged ports (10GbE/4/8Gb Fibre Channel). The 10GbE and 40GbE ports can be used as Ethernet data ports, Intelligent Resilient Framework (IRF) ports, or FCoE ports. The converged ports can be used as Ethernet data ports, IRF ports, or 4 or 8Gb native Fibre Channel ports. IRF support—An IRF domain simplifies management by combining multiple physical switches into one virtual switch. An IRF fabric appears as one node or virtual switch and is accessible as a single IP address on the network. You can use this IP address to log in at any member device to manage all the members of the IRF domain.

HPE Flex-10 or FlexFabric design When choosing between Flex-10 and FlexFabric, the first question to ask is whether the customer requires Fibre Channel connectivity. The key difference between Flex-10 and FlexFabric is that FlexFabric modules, shown in Figure 5-16, leverage the built-in converged network adapter in HPE BladeSystem servers to provide FCoE or integrated Small Computer System Interface (iSCSI) connectivity.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 5-16 HPE FlexFabric design FlexFabric and Flex-10 modules, the portfolio is shown in Figure 5-17, both provide iSCSI support, but FlexFabric provides FCoE support and Flex-10 does not. FlexFabric provides FCoE connectivity through the integrated converged network adapter and the FlexFabric modules. No additional components, such as a traditional HBA, are required.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-17 HPE BladeSystem Virtual Connect portfolio With the release of Virtual Connect firmware 4.01, Flex-10 and FlexFabric modules can also be used to provide dual-hop FCoE connectivity to a switch that supports FCoE connections, in which case the FCoE traffic traverses the Ethernet uplinks and connects to the SAN through the ToR or core switch. iSCSI support is provided through either FlexFabric or Flex-10 modules. If Flex-10 modules are used with Virtual Connect Fibre Channel modules, ensure an HBA is installed in the appropriate mezzanine slot in the blade. Then, you can simply configure a Fibre Channel HBA within the server profile and map it to the appropriate Fibre Channel SAN fabrics. In this case, FCoE SAN fabrics and FCoE FlexFabric adapters would not be used for Fibre Channel connectivity. Note Dual-hop FCoE connectivity is provided through Flex-10/10D, FlexFabric 10Gb/24-Port, and FlexFabric-20/40 F8 modules only. The original Flex-10 module does not support dual-hop FCoE. Fibre Channel connectivity is not available for Flex-10 modules. After you deploy the configuration, Fibre Channel connectivity could be added later by using Virtual Connect Fibre Channel modules and Fibre Channel HBAs. VMware ESXi is fully supported with BladeSystem and Virtual Connect. However, it is important to ensure that the proper network adapter and HBA drivers and firmware are properly installed in an ESXi environment.

Considerations when designing a FlexFabric switch environment

Technet24

||||||||||||||||||||

||||||||||||||||||||

When you are designing a storage solution, you need to consider the infrastructure, relevant storage components, and steps to move the design forward. Specific factors need to be considered at the customer site before implementation. When designing a FlexFabric switch environment, answer these questions about the customer environment before designing the solution: Is there Ethernet/FCoE support on 10GbE converged ports or 40GbE ports? Is there Fibre Channel 8 Gb, 4 Gb, or 2 Gb support on converged ports? Should the environment support Data Center Bridging (DCB) or FCoE-FCF? Is there FCoE or Fibre Channel NPV (N_Port Virtualization) gateway support? Does the customer have dual-hop support with HPE Virtual Connect blade switches and the HPE 6125XLG Ethernet Blade Switch? Does the customer have multihop support using FCoE VE_Port (seven hops) or Fibre Channel E_Port (three hops) ISLs? Does the environment have FlexFabric 5900CP series IRF support (up to nine switches with Ethernet, two switches with storage)? Is there iSCSI support? Does the environment use front-to-back or back-to-front airflow?

HPE Virtual Connect for Synergy HPE Virtual Connect Modules are interconnects with wire-once, change-ready templates that allow workloads to be moved without modifying the network. The technology is available for both HPE BladeSystem and HPE Synergy. HPE Virtual Connect Modules provide simple, seamless connectivity between virtualized compute modules and data center fabrics. The highly available, disaggregated architecture provides cost-efficient, easy-to-manage, and resilient fabric for traditional and nextgeneration workloads. The HPE Virtual Connect SE 40 GB F8 Module, shown in Figure 5-18, is ideal for environments requiring native FC, FCoE, and Flat SAN compute and storage connectivity with the Virtual Connect experience.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-18 HPE Virtual Connect for Synergy

HPE Synergy composable fabric

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 5-19 HPE Synergy composable fabric Up to 95% of network sprawl can be eliminated at the compute module edge with one device that converges traffic inside frames and directly connects to external LANs and SANs. HPE Synergy Composable Fabric’s disaggregated, rack-scale design uses a master/satellite architecture to consolidate data center network connections, reduce hardware and management complexity, and scale network bandwidth across multiple frames. The master module contains intelligent networking capabilities that extend connectivity to satellite frames through the HPE Synergy Interconnect Link Modules. This eliminates

||||||||||||||||||||

||||||||||||||||||||

the need for a top-of-rack (ToR) switch and substantially reduces cost. The reduction in components also simplifies fabric management at scale while consuming fewer ports at the data center aggregation layer. HPE Synergy Composable Fabric modules provide up to three redundant fabrics per frame, and the flexible QSFP (Quad Small Form-factor Pluggable) unified ports can be configured for either Ethernet or Fibre Channel. HPE Synergy Composable Fabric uses a flat, east/west architecture to maximize data throughput and minimize latency with only a single hop within large domains of virtual machines and up to 60 compute modules. By contrast, legacy hierarchical architectures use a north/south design that creates oversubscription bottlenecks and adds latency caused by multiple hops, both of which negatively impact performance.

HPE StoreFabric M-Series Ethernet switches—Optimized for storage HPE StoreFabric M-series Ethernet switch family, shown in Figure 5-20, offers industry-leading, high bandwidth and the lowest latency for a cost-effective solution to connect primary, secondary, hyperconverged, NAS, or object storage systems. With cost-effective options such as pay-as-you-grow licensing, seamless speed transition from 1GbE to 100GbE, and support for next-generation storage such as softwaredefined, scale-out, distributed, and NVMe (Non-Volatile Memory Express, ensure your M-series fabrics allow for implementing significant speed and architecture upgrades over time.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 5-20 HPE StoreFabric M-Series Ethernet switches are optimized for storage StoreFabric M-series can be tailored for specialized and critical workloads including exceptional performance in iSCSI storage and Remote Direct Memory Access (RDMA) server connected environments. HPE StoreFabric M-series provides a high-performance Ethernet Storage Fabric solution to support block, file, and object storage, thereby allowing you to build a converged network capable of simultaneously handling compute and storage traffic from iSCSI storage, to traditional arrays, to the newest NVMe over Fabrics arrays. It has unique storage-aware features such as DCB, which includes ETS (Enhanced Transmission Specification) and PFC (Priority Flow Control). It supports iSCSI TLVs, allowing iSCSI traffic to be specifically monitored and managed. The switches are upgradable in speed (up to 100Gb/s per port), port count, and software features, enabling a storage network ready for expansion, faster storage, faster servers, and expanded storage functionality from HPE. This combination of features, performance, flexibility, and storage-specific features makes HPE StoreFabric M-series the perfect solution for both small and large storage networks. HPE StoreFabric M-series SN2410M allows you to deploy high-performance Ethernet networking solutions that replace expensive, underperforming platforms. Providing industry-leading cut-through latency for any packet size and mix of port

||||||||||||||||||||

||||||||||||||||||||

speeds enables HPE. HPE StoreFabric M-series SN2700M Ethernet Switch offers a 100GbE-based network platform capable of delivering predictable performance and agility to keep pace with intense storage workloads small- to large-scale enterprise can produce.

HPE FlexFabric solution product portfolio Figure 5-21 shows the HPE FlexFabric portfolio. HPE 5930/5940 switches are modular DCB, iSCSI, and FCoE switches. HPE 5900CP is a very attractive switch platform supporting converged LAN and SAN (FC, FCoE, and iSCSI).

Figure 5-21 HPE FlexFabric solution product portfolio

HPE FlexFabric switches Considerations when designing a FlexFabric switch environment: Is there Ethernet/FCoE support on 10GbE converged ports or 40GbE ports? Is there Fibre Channel 8Gb, 4Gb, or 2Gb support on converged ports? Should the environment support DCB or FCoE-FCF? Is there FCoE or Fibre Channel NPV gateway support? For instance, the HPE FlexFabric 5940 Switch Series enables customers to scale their

Technet24

||||||||||||||||||||

||||||||||||||||||||

server-edge 10/40/100GbE ToR deployments with high-density 48 x 10GbE (SFP or BASE-T) with 6 x 40GbE ports, 48 x 10GbE (SFP or BASE-T) with 6 x 100GbE ports, and 32 x 40GbE ports, delivered in a 1RU design. It is also available with a two-slot modular version delivering two 40GbE ports and a four-slot modular version with four 40GbE ports. HPE FlexFabric 5940 Switch Series is a family of high-performance and low-latency 10GbE and 40GbE ToR data center switches. The 5940 Switch includes 100G uplink technology, which is part of the HPE FlexFabric data center solution and is a cornerstone of the FlexNetwork architecture.

||||||||||||||||||||

||||||||||||||||||||

Learning check StoreFabric M-Series switch performance scales from 1Gbe to 100GbE. True False

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Understanding SAN extensions Instructions: If you have not already done so, download the “HPE SAN Design Reference Guide” at: https://support.hpe.com/hpsc/doc/public/display?docId=c00403562&lang=enus&cc=us. On the left navigation pane, click SAN extension and bridging. When that page appears, click SAN extension. This should bring you to page 281. Write down the distances for each FC bandwidth noted as an extended ISL “Any distance greater than.”

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Configuring the most resilient HPE 3PAR configuration Figure 5-22 shows steps in 3PAR Configuration process: Determine proper array model. Add resiliency features. Add multisite capabilities.

Figure 5-22 Configuring the most resilient HPE 3PAR configuration

Determine proper array model: What type of provisioning: FPVV, TPVV, TPVV deduped, TPVV compressed, or TPVV DECO (deduplication and compression)? What are the expected data reduction ratios based on proper analysis, including data types? What are the random or sequential/block size, read write ratio, and response time expectations? What is the initial capacity requirement and the yearly growth? What is the initial performance requirement in IOPS, MBs, and response time plus the yearly growth? Set the compaction ratio expectation using the NinjaSTARS thin scan, NinjaCrawler, Get Thinner, and Adaptive Data Reduction (ADR) spreadsheet.

||||||||||||||||||||

||||||||||||||||||||

Choose a four-node system and one ending in a 40 or 50 model number for the extra CPU/cache.

Add resiliency features Change to four nodes to increase performance under failure/upgrade and avoid writethrough mode. Change to HA enclosure to protect against enclosure failure. Add eight drives to even the drive count per cage and avoid unusable space due to HAenclosure. Understand the pros and cons of HA enclosure regarding the additional availability versus the effects of large SSDs. Check planned initiator count against per port, per node, and per system limits in SPOCK (Single Point of Connectivity Knowledge). Check port count against performance limits per port and adjust for persistent ports/replication. Check spare space to see if you have enough space to rebuild under a failed enclosure. Check performance overhead for use of advanced software features in addition to compression already added. Check whether changing to a smaller drive can improve performance from the node count installed. Make sure to order the direct connect version for the enclosures.

Reach the next level with multisite capabilities: Add the All-Inclusive Multi-System Software license to enable replication (Remote Copy) and automated failover (Peer Persistence) if needed. Add the Data at Rest encryption license if needed and change to FIPS (Federal Information Processing Standards) drives. Add StoreOnce if needed as a backup target for RMC included in the all-inclusive basesystem license. Add a target 3PAR system if needed for replication, choose the mode, and size the link. If zero RPO/zero RTO (Recovery Point Objective/Recovery Time Objective) with automated failover is desired (Peer Persistence), check on additional link, cluster, and

Technet24

||||||||||||||||||||

||||||||||||||||||||

quorum witness requirements. If end-to-end Persistent Checksum is desired, check SPOCK for supported operating systems and HBAs. Choose the proper level of support based on customer needs. Be sure customer understands the benefit of setting up and paying attention to the call home functionality. Check on the environmental requirements where the arrays will be installed.

||||||||||||||||||||

||||||||||||||||||||

NinjaSTARS: Elements to be considered NinjaSTARS is HPE’s tool for sizing a 3PAR system; elements to be considered while using the tool are shown in Figure 5-23. It is available to both HPE internals and partners. It can be downloaded from the SAF (Storage Assessment Foundry) portal (https://saf.itcs.hpe.com). When you are sitting down with your customer to discuss the 3PAR configuration that best meets his needs, it is important to be clear about the workload the he will be migrating to 3PAR. Note To download NinjaSTARS, log in to the HPE Sizing Tools: https://saf.itcs.hpe.com/safdesktop/info/downloads

Figure 5-23 NinjaSTARS: Elements to be considered Performance derived from NinjaSTARS is based on real empirical data. There is a performance measurement database called PEM (performance management), which is based on performance runs from different combinations of workload types and 3PAR configurations. The performance information in the database is measured at the host, not at the 3PAR array. So when you look at the latency and IOPS numbers in NinjaSTARS, what you are seeing is front-end performance, not the backend performance to the drives, that is important to keep in mind.

Technet24

||||||||||||||||||||

NinjaSTARS sizing Support for 9450 and 20000 R2 models. PEM Hot Fix for performance data issue on 3PAR 8xxx and 20xxx with large block random workloads. Improved performance modeling for 3PAR ADR. New 3PAR ADR Calculator to estimate the benefits of deduplication and compression. New efficiency ratios based on lab testing and early access customer results. Ratios will be changed periodically based on customer experience/call home data. Supports DEDUP, COMPRESSION, and DECO.

Adaptive Data Reduction Please be aware that starting with version 2.9.7, NinjaSTARS includes modeling for 3PAR ADR. There is a new Flash option in the Ribbon menu, shown in Figure 5-24, which allows you to set the percentage of thin, deduped, compressed, or DECO volumes. For deduped and compressed workloads, we have de-rated the IOPS, throughput, and latency based on extensive tests by HPE Engineering. By default, 50% of the SSD capacity is set to use ADR. So to get the Ninja’s performance to base levels, the data reduction must be set to 0%. Options include: Data reduction ratio—Specify the data reduction as a ratio or in percent or use ratio calculator to determine expected benefits for known workloads. Weight—Specify the weights in percent of SSD workloads that will be using ADR. Reduction types –

Deduplication

–

Compression

–

DECO

–

Thin

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Figure 5-24 Adaptive Data Reduction

HPE 3PAR StoreServ 9450 hardware building blocks The building blocks of the 9450, shown in Figure 5-25, are very similar to the other 3PAR models. The backplane of the 9450 is 8u and can contain either two or four controller nodes, so in HPE’s configuration tools, you would select the base model, and then add the other two nodes if necessary. Next, you need to choose which host facing HBAs should be added to each node pair. The controllers on the 9450 do not contain any drives, and so you must add drives in the drive-only enclosures.

Figure 5-25 HPE 3PAR StoreServ 9450 hardware building blocks Since the 9450 can have only SSDs, we only need to have SFF enclosures. The 9450 might be racked either in the factory in an HPE rack or onsite in a customer supplied or HPE rack. For the HPE racks that contain the controllers, you can choose either the

Technet24

||||||||||||||||||||

||||||||||||||||||||

1075mm or 1200mm racks, and we will cover those options more in a minute. The last component is the Service Processor, which is used only for remote maintenance and connectivity to HPE Customer Support. The 9450 supports either a dedicated physical SP or the virtual SP. The HPE 3PAR StoreServ 9000 Storage is an enterprise-class flash array that helps you consolidate primary storage workloads—for file, block, and object—without compromising performance, scalability, data services, or resiliency. This newest 3PAR model based on the proven 3PAR architecture is purpose built for all-flash consolidation, delivering the performance, simplicity, and agility needed to support your hybrid IT environment. HPE 3PAR StoreServ 9000 Storage is available in a single all-flash model, the 9450, that offers rich Tier-1 data services and quad-node resiliency. Customers have the option of self-installing the HPE 3PAR StoreServ 9000 Storage array. The Customer Self Installation option is available for HPE 3PAR StoreServ 9000 Storage arrays that meet the following criteria: Two-node or four-node configurations CTO (configure-to-order) configurations (factory integrated) Single rack (the physical Service Processor can be in a separate rack) When performing an initial configuration of a StoreServ 9000 storage system, there are a few specifications to keep in mind:

HPE 3PAR StoreServ 9450 controller The LED on the front of the enclosure is used when servicing the 9450—blue means safe to remove and doubles up as a locate indicator, green is status ok, and amber indicates a fault. Figure 5-26 shows details of the HPE 3PAR StoreServ 9450.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-26 HPE 3PAR StoreServ 9450 controller

HPE 3PAR StoreServ 9000 configuration considerations HPE 3PAR StoreServ 9000 configurations start with the selection of the Base. The Base includes the controller node chassis and two controller nodes. Each Controller Node consists of CPUs, ASICs, Cache banks, and connectivity options for the storage array. HPE 3PAR 9450 two-node Storage Base with All-inclusive Single-system Software.

A minimum of one (1) storage base must be ordered for each array. Each storage base includes the controller node chassis and two (2) controller nodes. Each controller node includes two (2) SAS adapters and HPE 3PAR StoreServ 9000 Controller Nodes. Add a pair of the HPE 3PAR StoreServ 9000 Controller Nodes to the 9000 Storage Base to configure an HPE 3PAR StoreServ 9000 four-node array. HPE 3PAR 9450 Storage Node with All-inclusive Single-system Software

Each Controller Node SKU (stock keeping unit) represents a single (1) unit. Each Controller Node SKU includes two (2) SAS adapters. Controller Nodes are always ordered in pairs on an HPE 3PAR StoreServ 9000 Storage array. Only one (1) pair of 9450 Controller Nodes can be ordered per array. Technet24

||||||||||||||||||||

||||||||||||||||||||

Each Controller Node SKU is loaded with All-inclusive Single-system Software. The Controller Nodes have built-in Gigabit Ethernet ports for management and Remote Copy over IP. HPE 3PAR StoreServ 9000 Upgrade Controller Nodes

Use the HPE 3PAR StoreServ 9000 Upgrade Controller Nodes to increase the number of controller nodes on an existing previously installed HPE 3PAR StoreServ 9000 array. Increasing the controller nodes on an existing array helps improve resiliency, performance, and increase scalability. HPE 3PAR 9450 Upgrade Node with All-inclusive Single-system Software

Each Upgrade Controller Node SKU represents a single (1) unit. Each Upgrade Controller Node SKU includes two (2) SAS adapters. Upgrade Controller Nodes are always ordered in pairs on an HPE 3PAR StoreServ 9000 Storage array. Only one (1) pair of 9450 Upgrade Controller Nodes can be ordered per array.

Choose drive and host adapter HPE 3PAR StoreServ 9000 Storage uses SAS drive adapters for backend drive connectivity, as shown in Figure 5-27. Two SAS adapters are included as part of each controller node. Each SAS adapter on a controller node can connect up to eight drive enclosures. HPE 3PAR StoreServ 9000 Storage arrays have 12Gbps SAS backend.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-27 SAS and host adapters HPE 3PAR 9000 four-port 12Gb SAS Host Bus Adapter Two (2) SAS adapters are included as part of each controller node. A third optional SAS adapter can be ordered per controller node.

Choose host adapter Host adapters can be ordered separately to be installed in the field or they can be factory configured into controller nodes. Host adapter cards provide the array with additional FC ports, with 10Gb/sec iSCSI ports, or 10Gb/sec Ethernet ports. The additional FC ports can be used for multiple purposes including connection to hosts and connection to other HPE 3PAR StoreServ Storage systems in a Remote Copy or Peer Motion relationship. The iSCSI ports permit host connection in iSCSI environments. The Ethernet ports can be used to natively host various File protocols and core file data services.

HPE 3PAR StoreServ 9450 port scalability HPE 3PAR StoreServ 9000 Host Adapters include: HPE 3PAR 9000 4-port 16Gb Fiber Channel Host Bus Adapter HPE 3PAR 9000 2-port 10Gb iSCSI Converged Network Adapter Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR 9000 2-port 10Gb Ethernet Host Bus Adapter Considerations include: A minimum of one (1) host adapter per node must be ordered. The 16Gb Fiber Channel Adapter includes (4) 16Gb/s shortwave FC SFP+; The 10Gb iSCSI Adapter includes (2) 10Gb/s shortwave SFP+. The 10Gb Ethernet Adapter includes (2) 10Gb/s SR SFP+. Each node in a node pair must have the same number and type of adapters. The 4 ports of the FC adapter can be individually configured to connect to a host or to a remote array in a Remote Copy configuration. Scalability information is shown in Figure 5-28.

Figure 5-28 HPE 3PAR StoreServ 9450 port scalability HPE implemented the 8000 and 20000 models with a new iSCSI host CNA that is different from the ones on the 3PAR 7000 and 10000. The following are performance numbers you can use for guidance on the new models. However, because the HBAs are different, they do not apply to the older 3PAR model arrays: One dual-port iSCSI card can do 90,000 8K IOPS. One dual-port iSCSI card can do 1,350 MB per second. From a response time stand point, iSCSI has some additional overhead vs. FC, and so you can expect iSCSI response time to be about 1.6 times the equivalent FC response

||||||||||||||||||||

||||||||||||||||||||

time, as shown in Figure 5-29.

Figure 5-29 iSCSI performance

Configuring the 8000, 9450, and 20000 for utmost performance Host and drive ports are mapped to the CPU cores that handle interrupts, so more ports and more cores used, better the performance is. Direct connect uses more CPU cores when the number of enclosures per node pair is < 8 Provides maximum throughput. Supports connectivity for up to 48 enclosures on an eight-node 20000 (12 per node pair). Maximum scalability of 1152 drives (288 per node pair). Preferred option as long as more capacity scaling will never be required. Amount of CPU cores in each controller node is shown in Figure 5-30.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 5-30 Number of CPU cores in each controller node Daisy chaining supports connectivity for the full number of 96 enclosures and 2304 drives on an eight-node 20000. This will not provide more capacity on the 9450 today but might in the future. Select this option if future plans require this scalability. If the same number of ports is used, performance is the same with daisy chain and Direct Connect.

Choose drive enclosures for HPE 3PAR 9000 Add drive enclosures, shown in Figure 5-31, to include drives in the configuration (the controller nodes do not include any drive). Drive enclosures can be ordered separately for installation in the field, or they can be factory configured in a rack. The HPE 3PAR StoreServ 9450 supports up to forty-eight (48) drive enclosures, only one type of drive enclosure: 2U, 24 SFF (2.5 inch) drives.

Figure 5-31 Choose drive enclosures for HPE 3PAR 9000 Drive enclosures for HPE 3PAR 9000: HPE 3PAR 9000 24-disk 2U SFF (2.5in) SAS Drive Enclosure HPE 3PAR 9000 24-disk 2U SFF (2.5in) SAS Upgrade Drive Enclosure A minimum of 2 drive enclosures is required per node pair; each drive enclosure includes 24 drive bays for SFF 2.5” drives. Drive enclosures by default are connected directly to the controller nodes (direct

||||||||||||||||||||

||||||||||||||||||||

connect). As an option, it is possible to have the drive enclosures connected to the controller nodes in (two-level deep) daisy chains. The best practice is to balance the drive enclosures across the SAS ports on each controller node. With a four-node configuration, the best practice is to attach the same number of drive enclosures behind each node pair. Drive bays that are not filled with a drive must be covered with a drive blank to preserve proper air flow. If future capacity upgrades are expected, include enough drive enclosures so that there are some empty bays in each enclosure after all drives are added.

Choose drives options HPE 3PAR 9000 Drives are orderable at the time the array is purchased or can be added in the future when additional capacity is required. HPE 3PAR StoreServ 9000 drives are sold as single drives. HPE 3PAR SSDs

HPE 3PAR 9000 400GB SAS SFF (2.5in) SSD with All-inclusive Single-system Software HPE 3PAR 9000 1.92TB SAS SFF (2.5in) SSD with All-inclusive Single-system Software HPE 3PAR 9000 3.84TB SAS SFF (2.5in) SSD with All-inclusive Single-system Software HPE 3PAR 9000 7.68TB SAS SFF (2.5in) SSD with All-inclusive Single-system Software HPE 3PAR 9000 15.36TB SAS SFF (2.5in) SSD with All-inclusive Single-system Software HPE 3PAR FIPS Encrypted SSD

HPE 3PAR 9000 3.84TB SAS SFF (2.5in) FIPS Encrypted SSD with All-inclusive Single-system Software HPE 3PAR 9000 7.68TB SAS SFF (2.5in) FIPS Encrypted SSD with All-inclusive Single-system Software

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR 9000 15.36TB SAS SFF (2.5in) FIPS Encrypted SSD with all inclusive Single-system Software The minimum recommended initial quantity is eight (8) SSDs per node pair, and the minimum upgrade quantity is four drives per node pair or two drives per enclosure, whichever is larger. The best practice is to run Autonomic Rebalance (also known as tunesys) after adding the drives. RAID 6 is recommended, and all drive enclosures must contain an even number of drives, with a minimum of two. The best practice is to add equal numbers of drives to all enclosures, and in four-node configurations, the best practice is to attach the same number of drives to each node pair. SFF drives must be loaded in pairs of identical drives, beginning with the leftmost slot, slot 0, and filling to the right, leaving no empty slots between drives.

Drive slotting Start populating drives from left to right starting with the lowest row first and moving up, as shown in Figure 5-32.

Figure 5-32 Drive slotting Rules include: All drive enclosures must contain an even number of drives, with a minimum of two. A minimum of two drive enclosures is required per node pair. Best practices include: Configure multiples of four drive enclosures per node pair and HA cage. Maintain a balanced configuration across node pairs in terms of cards, drive enclosures, drives, and so on.

Drive enclosure connectivity

||||||||||||||||||||

||||||||||||||||||||

There are two options to connect the drive enclosures to the nodes: Direct connect for maximum performance

Uses more CPU cores when the number of enclosures per node pair is < 8. Provides maximum throughput. Supports connectivity for up to 48 enclosures on an eight-node 20000 (12 per node pair). Maximum scalability of 1152 drives (288 per node pair). Preferred option as long as more capacity scaling will never be required. Daisy Chain for maximum capacity (20000)

Connectivity for the full number of 96 enclosures and 2304 drives on an eight-node 20000. This will not provide more capacity on the 9450 today but might in the future. Select this option if future plans require this scalability.

9450 Drive Enclosure daisy chaining With daisy chaining, two drive enclosures are daisy chained behind a pair of SAS ports, as shown in Figure 5-33. This is to accommodate future drive scalability increase.

Figure 5-33 9450 Drive Enclosure daisy chaining

Technet24

||||||||||||||||||||

||||||||||||||||||||

9450 Drive Enclosure direct connect With direct connect, each drive enclosure is directly connected to the node SAS ports, as shown in Figure 5-34. This provides better performance and balanced configuration.

Figure 5-34 9450 Drive Enclosure direct connect

Choose cables Cables are required on the HPE 3PAR StoreServ 9000 Storage for drive enclosure connections and for host connectivity. Copper SAS cables are required for connecting

||||||||||||||||||||

the drive enclosures to the nodes on the base rack and for daisy chaining between adjacent drive enclosures. Active Optical SAS Cables are required for connecting drive enclosures in adjacent racks to the nodes in the base rack. OM4 Fiber Cables are required for host connectivity, Remote Copy, and Peer Motion. Cables include: SAS Active Optical Cables HPE External 0.5m (1ft) Mini-SAS HD 4x to Mini-SAS HD 4x Cable HPE External 2.0m (6ft) Mini-SAS HD 4x to Mini-SAS HD 4x Cable HPE 10m Mini SAS High Density Active Optical Cable HPE 25m Mini SAS High Density Active Optical Cable HPE 100m Mini SAS High Density Active Optical Cable HPE Premier Flex LC/LC Multi-mode OM4 2 fiber 5m Cable

Choose Service Processor The HPE 3PAR Service Processor remotely monitors the HPE 3PAR StoreServ 9000 and enables remote servicing of the array. The key capabilities of the Service Processor are to: Enable rapid, proactive responses to issues. Provide secure communication between the customer’s data center and HPE 3PAR Central for: –

Remote Online Software Upgrade—Upgrade software with no application disruption

–

Remote Diagnostics—Maintain key diagnostic information centrally on a historical basis

–

Remote Serviceability—Provide fast predictive response and remediation

||||||||||||||||||||

Each HPE 3PAR StoreServ 9000 requires its own Service Processor. The Service Processor functions as the communication interface between a customer’s IP network and HPE 3PAR Central by managing all service-related communications. The Service Processor leverages the industry-standard HTTP over Secure Sockets Layer (HTTPS) protocol to secure and encrypt data communication. The Service Processor can be deployed either as a virtual Service Processor (VSP) or a physical Service Processor.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Virtual Service Processor

A VSP is included free with the base HPE 3PAR Operating System. The VSP can be installed on a customer-provided VMware or Microsoft Hyper-V system. Physical Service Processor

The physical Service Processor is a dedicated storage appliance located within the storage rack providing close proximity to the HPE 3PAR StoreServ 9000 Storage. The physical Service Processor is fully supported and maintained by HPE Services. The physical Service Processor has serial port connectivity that provides maintenance access for trouble shooting capabilities.

HPE 3PAR StoreServ 9450 racking options The HPE 3PAR StoreServ 9000 Storage is available factory integrated in the HPE Intelligent Series racks with the appropriate power distribution units (PDUs), as shown in Figure 5-35. Alternatively, a third party rackmount kit can be purchased in cases where the HPE 3PAR StoreServ 9000 Storage needs to be installed on any rack other than the HPE racks HPE 42U 600mm x 1075mm Enterprise Shock Rack. When installed in a 42U 1075mm rack (BW904A), the array uses horizontal modular PDUs and can support up to 12 drive enclosures in the base rack. When installed in a 42U 1200mm rack (BW908A) the array uses vertical PDUs and can support up to 16 drive enclosures in the base rack. Expansion racks can support up to 20 drive enclosures. Note The Third Party Rack Mount Kit is required for mounting the 3PAR StoreServ 9000 Node enclosure in third-party racks. For mounting the drive enclosure, rail kits are included with the product.

Note For 1200mm deep, factory-integrated racks, six power connections are required for US/Japan Single Phase PDUs. International single-phase and all three-phase power solutions require four power connections per rack.

||||||||||||||||||||

||||||||||||||||||||

Note For more information on the HPE rack offerings, please see the following URL: http://h18004.www1.hpe.com/products/servers/platforms/rackandpower.html For more information on rack options, see: http://www.hpe.com/products/rackoptions

Figure 5-35 HPE 3PAR StoreServ 9450 racking options

HPE 3PAR StoreServ 9450 warranty, support, and installation Warranty: 24x7 four-hour onsite response Duration: Five years for all SSDs, three years for the array and all hardware components (except SSDs) Support: Foundation Care, Proactive Care (default), Proactive Care Advanced, Datacenter Care Support SKUs similar to other HPE 3PAR families Seven-year support for covering media wear-out and electronic failure Installation and Startup Services: Structure similar to HPE 3PAR StoreServ 8000

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ 9450 performance Figure 5-36 shows preliminary numbers from empirical measurements and estimation as of June 2017. Final numbers are available in NinjaSTARS.

Figure 5-36 HPE 3PAR StoreServ 9450 performance

||||||||||||||||||||

||||||||||||||||||||

Learning check When discussing response times with your customer, which solution would typically offer the lowest: ISCSI or Fibre Channel?

Technet24

||||||||||||||||||||

Activity: Using NinjaSTARS to create a configuration Instructions: Log in to the HPE Sizing Tools at: https://saf.itcs.hpe.com/safdesktop/info/downloads/ for this activity. Using NinjaSTARS, create a configuration for a customer who requires: –

60 TB (57.9 TB) of usable capacity on their 2 x node 3PAR StoreServ 9450 with virtual Service Processor in a 42U rack.

–

The configuration will be HA cage with 24 x 3.84 TB SSD disks, 8 x 16 GB FC ports, and 4 x 10 GB CNA ports.

–

16 TB for VMs (virtual machines).

–

12 TB for Oracle database.

–

10 TB for Exchange.

–

12 TB for other applications.

–

10 TB for future growth.

–

Data is typically 8k random 70-30 mix.

–

Using RAID 6 6+2.

–

24 x 3.84 TB SSD disks.

||||||||||||||||||||

Notes:

||||||||||||||||||||

||||||||||||||||||||

Examples from NinjaSTARS Figures 5-37, 5-38, 5-39, and 5-40 show an example of using NinjaSTARS.

Figure 5-37 Configuration tab

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 5-38 Performance tab

||||||||||||||||||||

||||||||||||||||||||

Figure 5-39 Adaptive Data Reduction Calculator

Figure 5-40 Bill of Materials

Technet24

||||||||||||||||||||

Activity: Using Nimble Config Sizer to create a configuration Instructions: Log in to the HPE Sizing Tools at: https://saf.itcs.hpe.com/safdesktop/info/downloads/ for this activity. Using Nimble Config Sizer create a configuration for a customer who requires: –

90 TB of usable capacity for a VDI environment

–

Size model to meet application needs of 120,000 IOPS

–

Retain seven days of snapshots

–

Future proof for two years of growth

–

Set network requirements for 10 Gb iSCSI SFP+ 2 port card

||||||||||||||||||||

Download the sizing tool, and create the sizing solution to meet the customer requirements. Use the “Data Reduction” chart to gauge the estimated data reduction by application (for example, VDI is around 7.33:1 reduction). Copy the Letter of Understanding to the clipboard, and save to a Word document. Notes:

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE Reference Architectures HPE works to profile and understand many common IT solutions from a performance, scalability, and applicability perspective. The results are documents that function as blueprints, detailing specific configurations and delivering operational characteristics defined as reference architectures. Reference architectures, example shown in Figure 5-41, are documents or templates that outline an infrastructure configuration specifically designed for integration that details the required infrastructure components and method of assembly. They explain how to build a complete solution based on leading, modular, standards-based technologies designed for convergence and include a fair amount of flexibility in the configuration.

Figure 5-41 Example of HPE Reference Architecture HPE Reference Architectures and HPE Reference Configurations include complete configuration, sizing, bill-of-material, and deployment details. We deliver tested recipes for success for cloud, client virtualizations, databases, Dev/Ops, collaboration, and more—including validation of third-party hardware and software, as shown in Figure 542.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-42 HPE Reference Architectures HPE provides a comprehensive set of reference architectures that allow customers to design storage solutions easily and cost effectively. These reference architectures include IT building blocks that are assembled to address key enterprise workloads including: Cloud Client Virtualization Big Data Data Management Business Continuity Collaboration

HPE Converged Architecture 700 with Arista

Technet24

||||||||||||||||||||

||||||||||||||||||||

Executive summary HPE Converged Architecture 700 with Arista delivers a scalable, converged infrastructure platform for virtualization that provides tuned infrastructure to run your workloads and applications. These solutions are delivered through certified channel partners and provide infrastructure your way, delivered as one integrated stack, saving your time and resources. HPE Converged Architecture 700 with Arista includes HPE lab-tested infrastructure, which combines Arista network switches, HPE industry-leading x86 servers, HPE 3PAR StoreServ Storage, and the VMware® vSphere hypervisor. The various sections of the Reference Architecture, shown in Figure 5-43, identify steps to configuring based on best practices. For example, Solution design and best practices implementation section lays out the required and optional features for each component of the solution and how to best configure them. It also describes how the solution achieves the goals of manageability, security, performance, and integration in the larger data center environment. Note Go to the HPE Enterprise Library for more information: http://h17007.www1.hpe.com/us/en/enterprise/reference-architecture/infolibrary/index.aspx

||||||||||||||||||||

||||||||||||||||||||

Figure 5-43 HPE Converged Architecture 700 with Arista

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Reference Architecture Instructions: Log in to the HPE Information Library at: www.hpe.com/info/enterprise/docs for this activity. Using the Reference Architecture for HPE Converged Architecture 700 with Arista VMware 6.0 design, list the Storage BOM (bill of materials) for this design.

Figure 5-44 shows a table from HPE Reference Architectures containing Bill of materials.

||||||||||||||||||||

||||||||||||||||||||

Figure 5-44 HPE Reference Architectures: BOM

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check Write a summary of the key concepts presented in this chapter.

||||||||||||||||||||

||||||||||||||||||||

Summary Planning and designing an enterprise storage system that can handle large volumes of data and a large number of users is a multistep process with many decision points along the way. The RAID configuration you choose for the storage system depends on the customer’s plans for data fault tolerance, data availability, and capacity growth. SANs provide flexibility in system management, configuration, connectivity, and performance to meet the needs of the changing business environment. When performing an initial configuration of a StoreServ 8000 storage system, there are certain factors to consider such as RAID configurations that permit HA cage. The key difference between Flex-10 and FlexFabric is that FlexFabric modules leverage the built-in converged network adapter in BladeSystem servers to provide FCoE or iSCSI connectivity. HPE Reference Architectures include IT building blocks that are assembled to address key enterprise workloads.

Technet24

||||||||||||||||||||

||||||||||||||||||||

6

Staging an Effective Storage Consulting Engagement

||||||||||||||||||||

||||||||||||||||||||

OBJECTIVES After completing this chapter, you should be able to: Describe the goals of Hewlett Packard Enterprise (HPE) presales consultants and storage architects. Identify consultative selling. Explain how to use HPE sizing and planning tools during a consulting engagement. List the available return on investment (ROI) and total cost of ownership (TCO) tools. Direct others to HPE resources online. List steps for a Proof-of-Concept (PoC) test.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Prelearning check You are looking to book a meeting to discuss a project on which you are working. With whom should you be discussing business goals and objectives?

||||||||||||||||||||

||||||||||||||||||||

Presales consulting goals As an HPE presales consultant or partner, your primary goal is to find and convey value to your customers. You need to learn as much as you can about the customer’s business situation and needs, as shown in Figure 6-1. To fully understanding their needs, you should be able to: Communicate value by providing business insights. Identify other ways to support the customer through cross-selling. Develop and maintain long-term relationships that can lead to more business. Distinguish HPE from its competitors. Be the customer’s first call when they need support.

Figure 6-1 Presales consulting goals

Technet24

||||||||||||||||||||

||||||||||||||||||||

It is so important that we make every single customer interaction as valuable as it can be. We can do this by knowing our customers, understanding our values, making ourselves relevant to our customers, and delivering meaningful business outcomes.

Customer culture Customer culture is fundamental to HPE success, so let us be much more properly focused on customers. Let us make it real and do something customer related every single day. By doing this, we can achieve something incredibly powerful.

||||||||||||||||||||

||||||||||||||||||||

Recognize the buyer, their top priority, and desired outcomes IT spending by non-IT business units is growing more than twice as fast as traditional IT departments. A long time ago, IT had a special place in most organizations. Working under G&A (General and Administrative Expenses) or finance, or reporting directly to the CEO, IT was the group that other departments had to petition to find appropriate technology solutions for their business needs. However, according to a new report commissioned by HPE and conducted by C Space, this model is becoming increasingly outdated as lines of business (LoBs) become more involved in critical IT decision making. Discussions with hundreds of IT and business managers, as shown in Figure 6-2, found that for some LoB managers, more than 30% of their time is spent making IT-related decisions. This includes identifying IT needs, realizing opportunities to use technology to overcome inefficiencies or improve operations, researching available solutions, and keeping up to date on technology trends.

Figure 6-2 Recognize the buyer, their top priority, and desired outcomes Recognize the buyer, their top priority, and desired outcomes are important before you decide which storage solution fits and meets the buyer needs. The HPE 3PAR StoreServ portfolio is the choice for many storage savvy buyers who are looking for great features and flexibilities in the Tier-1 storage space, primarily focusing on data center solutions with great disaster tolerance, protection, and highly Technet24

||||||||||||||||||||

||||||||||||||||||||

available.

Align the right solution by workload and use case Regardless of who makes the IT purchasing decision, IT spending by non-IT business units will reach $609 billion in 2017, according to an IDC (International Data Corporation—a market research company), forecast, and it will continue to grow at a 5.9% compound annual growth rate (CAGR) through 2020. This is more than twice the 2.3% CAGR IDC predicts for technology spending by IT departments. This year, non-IT business units will spend a projected $150.7 billion on applications, $120.3 billion on project-oriented services, and $70.3 billion on outsourcing, according to IDC. Traditionally, expenditures on hardware, applications, and infrastructure have come out of the IT budget; however, as IT spending has remained flat relative to the overall budget, technology needs are increasing. LoBs are using their own business budgets and project budgets to pay for the technology, so project managers are approving purchases on their own, rather than going to IT to get approval. Note To read more information about the IDC forecast, “Technology Purchases from Line of Business Budgets Forecast to Grow Faster Than Purchases Funded by the IT Organization,” visit: https://www.idc.com/getdoc.jsp? containerId=prUS42405217. It is important to align the right solution by workload and use case, as shown in Figure 6-3. Even with self-selection, there is workload affinity to certain classes of infrastructure. For small deployments, consolidating a few applications with good price/performance, but limited service-level agreements (SLA) requirements, start with MSA2x00 and consider SimpliVity or Nimble for more data services. For VM/VDI (virtualization and virtual desktop infrastructure) environment, productivity and collaborations, small private cloud for VMs, or moderate Structured Query Language (SQL) database (DB) demands, start with SimpliVity and consider Nimble if your customer is open to new approach but has physical hosts and nonhyperconverged needs. For DBs, emerging applications, Storage-as-a-Service with VMs, Microsoft

||||||||||||||||||||

||||||||||||||||||||

applications, or container environments, start with Nimble and consider 3PAR for greater configurability, performance, and multi-tenant, multi-arrays scale. For a large private cloud, performance DBs, HPE Synergy-attach, ITaaS (IT-as-aService), VMs, and OLTP analytics, start with 3PAR and consider XP7 for extreme availability and/or mainframe connectivity.

Figure 6-3 Align the right solution by workload and use case

Use the flash conversation to identify data protection opportunities Recognize the buyer, their top priority, and desired outcomes, as shown in Figure 6-4. For Veeam backup customers using any primary storage array who want to run production workloads on that data (such asdev/test, QA (Quality assurance), analytics), start with Nimble SF (Nimble Secondary Flash Array). Nimble SF brings active backup and availability.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 6-4 Use the flash conversation to identify data protection opportunities For cost- and performance-optimized solution for data protection ingest and recovery, with extensive independent software vendor (ISV)/app support to consolidate backup from multiple sources, start with StoreOnce/Recovery Manager Central (RMC). StoreOnce and RMC is comprehensive and low-cost backup solution.

||||||||||||||||||||

||||||||||||||||||||

Do not miss the secondary storage opportunity Do not miss the secondary storage opportunity, as shown in Figure 6-5. Selling MSA? Sell StoreOnce single-node solutions. Low-cost solution with extended ISV ecosystem to protect data and consider StoreEver tape devices or JBODs (Just Bunch of Disks) for lower-cost options. Selling SimpliVity? Leverage built in SimpliVity backup, VM-centric protection to get the most from SimpliVity investments. Consider StoreOnce as a lower-cost option or for heterogeneous environment. Combine with Veeam for granular app integration and recovery. Selling Nimble? Sell Secondary Flash with Veeam Software. Affordable for backup, but fast enough for running workloads such as dev/test, QA, analytics. Consider StoreOnce as a lower-cost option or for heterogeneous environments. Selling 3PAR? Sell StoreOnce with RMC/Veeam. Cost-effective, dedicated dedupe target for snaps and backups across many apps/ISVs. Consider StoreOnce and Data Protector or Veritas to protect non–VM-based and enterprise environments.

Figure 6-5 Do not miss the secondary storage opportunity Differentiate with integrated availability, performance, backup, disaster recovery, and more.

Consulting engagement stages There are five stages to the consulting process, as shown in Figure 6-6. Keep these notes in mind as you progress through each stage: Prepare

Technet24

||||||||||||||||||||

–

Learn as much as possible before you begin to engage a customer.

–

Research the company and the people, find out about areas of concern, and decide who is the best person to contact.

Interview –

Know what you want to learn before the initial interview with the customer.

–

Consider the person’s position at the company and anticipate their concerns accordingly.

–

Listen to what the customer has to say.

–

Use time effectively and address their concerns in a timely manner.

Plan –

Determine which solutions are closely aligned with the customer’s business needs.

–

Plan the next steps in the project, including who to add to your team when the time comes.

Propose –

Design the solution that meets the customer’s needs and successfully conveys value.

–

Do not propose products only; recommend a complete solution that demonstrates value at all levels of the organization.

Present –

Articulate the recommended solution, positioning it in terms that the customer understands.

–

Ensure that the customer knows HPE has the solutions to meet their needs.

||||||||||||||||||||

Figure 6-6 Consulting engagement stages

Where do projects come from? IT is considered at every level of the organization, as shown in Figure 6-7. Presales

||||||||||||||||||||

||||||||||||||||||||

consultants should guide customers through each business impact while understanding the following: Their Business Goals Their Market Conditions Their Overall Strategy Their Challenges Their Motivation

Figure 6-7 Where do projects come from?

Learning check You are looking to book a meeting to discuss a project on which you are working. With whom should you be discussing business goals and objectives?

Technet24

||||||||||||||||||||

||||||||||||||||||||

Business value sales with storage To understand business value and value selling, this simple formula, shown in Figure 68, is a good starting point: Profits = Revenue – Expenses Demonstrating business value consists of creating additional profit for customers by influencing the components of this equation. This can mean a growth in revenue, a reduction in expenses, or both.

Figure 6-8 Business value sales with storage Value selling is the process of helping customers connect IT to their business and measuring the potential improvement in business value if they decide to change their IT strategies. This process leads to a much stronger customer relationship and, ultimately, increased sales. Key points to address are shown in Figure 6-9.

||||||||||||||||||||

||||||||||||||||||||

Figure 6-9 Key points to address IT has become a significant part of corporate spending. Large IT budgets require executives to be involved in IT spending decisions. Business executives focus on revenue, profitability, and ROI. To guide customers’ IT purchasing decisions, storage architects should assume the role of business consultant. To make the best decisions for their companies, business executives need to understand the impact of IT investments on their business. Rationalize IT investments relative to other strategic investments (such as research and development, plant, and equipment). Evaluate the financial case for investing in new technology.

Engage early in the sales cycle with consultative selling HPE Technology Consulting Services, shown in Figure 6-10, enable you to help your customers transform to the New Style of Business with confidence. With a proven track

Technet24

||||||||||||||||||||

||||||||||||||||||||

record of success, our Consulting Services portfolio helps you expand your services offerings and deliver a complete solution to our mutual customers. A wide portfolio of Services covering the modern IT trends not only helps you deliver on your customers’ present needs but also helps you enhance your expertise for the future.

Figure 6-10 Engage early in the sales cycle with consultative selling With HPE Technology Consulting Services, you can: Leverage HPE global expertise, IP, and resources to solve complex challenges. Expand your services offerings and enhance your status as a trusted IT advisor. Transform your clients’ environments to facilitate a New Style of Business. The HPE Technology Services value framework shows the lifecycle of services, how we partner with the customer to help their IT make the case for change for the business, show them how they can transform “safely” keeping the business stable.

HPE Storage business value methodology HPE Storage solutions improve operational efficiencies, service levels, and business

||||||||||||||||||||

flexibility to create business value, as shown in Figure 6-11. This methodology goes beyond simple TCO or ROI analysis. It captures both hard (cost savings) and soft (revenue-related) dollars generated by storage investments.

Figure 6-11 HPE Storage business value methodology

Defining business value in storage There are three contributors to business value in a storage context: Operational efficiencies –

Hardware, software, and network

–

Server utilization

–

Storage utilization

–

Backup costs

–

Asset management

–

Network costs

Business flexibility –

Time to market

–

Application deployment

–

Rate of application change

Service levels –

Availability

–

Performance

–

Number of users

–

Impact of downtime

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

These elements directly influence company revenues or costs, and each can be further characterized by specific subelements, such as staff utilization, number of users, availability, application deployment, and so on.

Addressing business objectives In the example shown in Figure 6-12, a business is focused on two goals to meet its overall strategy: grow revenue and profitability and expand its customer base.

Figure 6-12 Addressing business objectives IT capabilities can address these drivers in four ways: Streamlining business operations—Centralized IT and purchasing, integrated systems

||||||||||||||||||||

that reduce redundancy and lag time Accelerating time to market—24x7 availability for applications, DBs, and messaging systems; increased response times Exploring new opportunities—Use of data mining, customer relationship management, and integrated systems for cross-selling Overtaking the competition—More efficient cost structure, productivity tools, and services to enable new business opportunities

Steps to a successfully architected storage solution When collaborating with potential customers, it is important to clearly demonstrate your plan for developing and implementing architected storage solutions. To deliver a successful consulting engagement, follow these three overarching steps, as shown in Figure 6-13: Architect –

Create the concept that will meet (or exceed) the customer’s business needs today and tomorrow.

–

Develop a blueprint to help customers visualize the proposal. Include details about why you are recommending these solutions and what benefits they will provide.

Design –

Detail which solutions will be used and how they will be implemented, including what resources (tools, people, and equipment) will be necessary.

Articulate –

Ensure that all business executives who are involved in IT decision making can understand the proposal and see the benefits. When people want change, they will support the new solutions, which might help with the implementation, management, troubleshooting, and so on.

–

Change is not always a popular decision. It is important to demonstrate the benefits of any changes to get employees onboard.

||||||||||||||||||||

Technet24

||||||||||||||||||||

Figure 6-13 Steps to a successfully architected storage solution

Four fundamental storage architecture views There are four storage architecture perspectives that can help with your development, including business, functional, technical, and implementation views. Each view covers a different angle and includes questions the customer might want to know before agreeing to an HPE solution. Be prepared with answers and questions shown in Figure 6-14: Business view—Why? –

Why we need to upgrade our current infrastructure?

–

Why this is important to meet our business goals?

Storage functional view—What? –

What will the storage accomplish?

–

What will the overall organization be regarding storage operation, capabilities, and services?

–

What will the new infrastructure do for users?

–

What information will it provide to the business?

Storage technical view—How? –

How will the new infrastructure be realized from an overall technical point of view?

–

How will the applications, data, interfaces, and storage infrastructure interact?

Storage implementation view—With? –

Which specific products and other components will be used?

–

Which business departments will be affected by the implementation?

–

Are there any existing or potential technical and organizational constraints?

||||||||||||||||||||

||||||||||||||||||||

–

How will the new infrastructure be implemented and supported by IT staff?

–

What technology and product decisions need to be made?

||||||||||||||||||||

Figure 6-14 Four fundamental storage architecture views

Technet24

||||||||||||||||||||

HPE sizing and planning tools HPE helps partners and customers create the best solution for each IT problem by providing multiple sizing and planning tools. All of those tools are available without charge, but access to some tools might require registration. To use HPE sizing tools most effectively: Determine the customer’s business application requirements. Pay close attention to how much computing equipment is required. –

If overconfigured, the solution does not return good value.

–

If underconfigured, the solution does not provide adequate service.

–

Several hardware configurations might apply.

||||||||||||||||||||

Understand the hardware components and sizing requirements. Consider the price-to-performance ratio, budget, and time frame for the purchase. Create specifications for the configuration options. Identify service levels and how software and hardware will be used.

HPE Storage Sizing Tool The Storage Sizing Tool, shown in Figure 6-15, can be downloaded from the HPE Partner Portal; you will also need to downlaod the license as well or ask via emai. You will find all instructions and details on the HPE Partner Portal.

||||||||||||||||||||

||||||||||||||||||||

Figure 6-15 HPE Storage Sizing Tool In addition to sizing tools, HPE provides a number of additional presales tools and resources. The HPE Storage Sizer is a sizing tool that helps you design a storage infrastructure to meet the needs of a customer. The Storage Sizing Tool can be downloaded from the HPE website. This is an important feature because it keeps the sizer current and any configuration prepared using this tool will be a valid, fully supported configuration. The Storage Sizer supports the disk storage subsystem and other storage solutions such as backup systems, network-attached storage (NAS) solutions, and other storage components. The Storage Sizer requires a license. The HPE Storage Sizer provides the following features and benefits: Simplifies the process of designing a storage solution Applies storage design, licensing, and services rules Provides output as a valid, supported configuration that can be imported directly into SalesBuilder for Windows (SBW) for quotation Provides localized parts and pricing for different geographic regions Includes HPE Smart Update technology, which brings new products or functionality to you through an Internet connection Encompasses the HPE Storage family

Technet24

||||||||||||||||||||

||||||||||||||||||||

Initiates an update for every product launch as part of the new product introduction process Includes new functionality, which was added based on user input, annual surveys, and quarterly focus groups.

Using the HPE Storage Sizer The Storage Sizer enables you to work with your customers to design a storage infrastructure that will meet their online and offline needs. You can define customer requirements, such as: Performance requirements with specific metrics Business requirements, such as server consolidation Pure capacity requirements For example, additional requirements might include raw capacity, estimated IOPS, replication and backup criteria, and number of host ports. Because the tool applies all the HPE storage area network (SAN) design rules, it provides a valid, supported storage infrastructure to meet the requirements of your customer. Use the Storage Sizer when you are not sure which combination of products will best address customer requirements. This tool lets you try different solutions. A helpful wizard interface guides you through the process of sizing a SAN by asking a series of questions about the proposed configuration. This wizard is intended for those who have less experience using the Storage Sizer. Note To access the HPE Storage Sizing Tool, open the hyperlink: https://sizersllb.itcs.hpe.com/swdsizerweb/

||||||||||||||||||||

||||||||||||||||||||

HPE Nimble Config Sizer The HPE Nimble Configuration Sizer is a downloadable sizing tool that helps to design a storage infrastructure that will meet your customer needs. The Nimble Config Sizer tool allows you to enter information to correctly size an HPE Nimble Storage array. Tabs along the bottom give you valid configuration information along with SKUs (stock keeping units) and a change history. Always check for an updated version of the tool. Upgrades to existing Nimble Storage arrays are not handled by this spreadsheet—Contact [email protected] The Excel-Based tool will ultimately be absorbed into the HPE Ninja Storage Assessment Recommendation and Sizing (STARS) suite, planned for 2018. The HPE InfoSight website will soon have a web-based tool for more complex solution sizing with detailed performance sizing. This is also planned for 2018. Note Storage tools for sizing can be downloaded from: https://saf.itcs.hpe.com/safdesktop/info/downloads/ HPE product codes can be obtained as input into One Config Advanced (OCA).

Storage Assessment Foundry Storage Assessment Foundry, shown in Figure 6-16, is a new suite of tools for presales and internals only; it comprises two components that can be used for sizing: SAF (Storage Assessment Foundry) Collect and SAF Analyze.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 6-16 Storage Assessment Foundry The collector is a lightweight and agentless tool that would be run against a customer’s storage array to access the performance and configuration information from the array. Then, the data is securely uploaded to HPE and automatically analyzed. When that is complete, you can go to the SAF self-service portal, view the Summary and Drill-down into the analysis reports; this can be used to create a NinjaSTARS configuration. The collector currently supports HPE 3PAR and EMC VMAX and VNX arrays; additional third-party storage arrays are under development, and there is the potential to collect information from host systems as well as from the storage directly. After the data is analyzed, you will first see a summary screen. You can look at the information from a single storage array, or if your customer has, for example, multiple VNX storage arrays, you can combine them into a single view. The tool is interactive— you can drill down further into this information, or you can just use this summary to input directly into NinjaSTARS. Figure 6-17 shows an example of assessment summary screen; Figure 6-18 shows additional analysis and reports.

||||||||||||||||||||

||||||||||||||||||||

Figure 6-17 Assessment Summary

Figure 6-18 Examples of analysis and reports The collector must be run from a Command Prompt to be able to add the required arguments to the collector; however, it does not need to be run with admin rights (assuming the user is allowed to write to the current directory). To collect capacity and configuration data and/or performance statistics for HPE StoreServ 3PAR arrays, use SAFcollector SS3PAR with the required options, for example: C:\Storage Assessment Foundry>safcollector 3paradm -p 3pardata -i 15.ip.ip.183

SS3PAR

-u

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 6-19 shows an example of storage sales process.

Figure 6-19 Sales process Analytics for HPE Storage can be found at the Storage Assessment Foundry website https://saf.itcs.hpe.com. Website content is shown in Figure 6-20.

||||||||||||||||||||

||||||||||||||||||||

Figure 6-20 Storage Assessment Foundry Suite

Storage Assessment Foundry Suite NinjaSTARS for 3PAR and the Nimble Config Sizer are storage sizing tools for HPE Storage arrays. They can be used to size and configure cost-effective and performanceoptimized solutions for both HPE 3PAR and HPE Nimble Storage. Using a basic set of capacity and performance metrics (requirements) from an existing storage environment, you can use these tools to interactively change the requirements and see the change in capacity and performance characteristics of the target storage array. In addition, both tools provide guidance on the anticipated capacity reduction benefits of a given application profile on each storage array. The HPE Nimble Space Savings Estimator (NimbleSSE) is a command line utility that Technet24

||||||||||||||||||||

||||||||||||||||||||

provides an estimate for how much space a user could save on a disk or volume. The dataset can reside on any storage that is accessible to a host on which NimbleSSE is installed. The tool runs on a Windows host (either physical or a virtual machine), scans one or more disks or volumes, and calculates the amount of space a user might save by using deduplication, compression, and zero block pruning. NinjaCrawler is a tool that reads actual raw data on a host and determines the possible capacity savings that could be achieved with data deduplication and compression. NinjaCrawler relies on HPE 3PAR deduplication and compression technologies to emulate the amount of data reduction the specified volume(s) would benefit from if they were stored on a 3PAR array with those technologies enabled. NinjaCrawler works on both Linux and Windows platforms.

HPE NinjaSTARS for HPE 3PAR StoreServ HPE NinjaSTARS Tool, shown in Figure 6-21, is an interactive storage sizing tool for HPE 3PAR StoreServ Storage arrays. It can size and configure a cost-effective and performance-optimized HPE 3PAR StoreServ that can replace existing legacy arrays. Based on a set of basic capacity and performance metrics (requirements) from an existing storage environment, it converts these requirements into a ready-to-order bill of materials (BOM) list.

Figure 6-21 HPE NinjaSTARS for HPE 3PAR StoreServ

||||||||||||||||||||

||||||||||||||||||||

The tool has different ways of accepting the sizing requirement, including using a set of standard templates, inputting a desired usable capacity number (with percentages per tier), or completely customizing a StoreServ. Users can interactively change the requirements to see the change in capacity and performance characteristics of the target HPE 3PAR StoreServ. NinjaSTARS also provides options for licensing and software components that are offered for any HPE 3PAR StoreServ. NinjaSTARS uses a flow-based approach in the following sequence: Gather configuration and sizing requirements. Interactively customize the target HPE 3PAR StoreServ. Select licensing and software components for final configuration. Generate a BOM for the selected configuration.

Installing NinjaSTARS NinjaSTARS is an agentless, standalone tool, designed to run on any Windows machine, including a laptop, workstation, or server. It is compiled as a portable application, designed to run from a USB stick; no installation is required. Scanning of the target systems does not examine any individual file, require a host agent, or leave behind any information. Network traffic generated is negligible.

How to achieve better data protection How to achieve better data protection? Use HPE StoreOnce NinjaProtected backup assessment and Get Protected Guarantee program. Backup assessments have become one of the most popular storage services from HPE. The process takes less than an hour, depending on the size of your environment. We capture metadata from the backup application itself and thereby access no real customer data. Each assessment analyzes key industry backup metrics from all the backup jobs and provides backup window, backup success rate, failed backups, number of restores, failed restores, and potential saving through data deduplication. After we analyze the data, you receive an executive report with an overall assessment of your backup environment and a benchmark for any future assessments. LIGHTWEIGHT and NON-INVASIVE

Technet24

||||||||||||||||||||

Assessment tool is an agent-less, non-invasive script tool. No installation is required on the system. SIMPLE, POWERFUL, AND EFFICIENT Automates the gathering of existing backup environment data. Assesses overall backup health of the environment. Leverages historical stats from backup servers.

Get Protected Guarantee Program By performing a free NinjaProtected backup assessment, HPE and its partners can demonstrate how customers can reduce the amount of backup data they need to store by 95% as compared to a fully hydrated backup. Details

Applies to any new HPE StoreOnce System including StoreOnce VSA (virtual storage appliance) –

Data types include file servers, MS Exchange, and virtual machines (using agentless backup methods).

–

Audio and video files cannot exceed 10% of the total data within a backup.

||||||||||||||||||||

If the guarantee is not met, HPE will make up the difference with additional capacity and support. GPG program requirements

Complete the NinjaProtected assessment to determine whether you qualify for the GPG. Backup applications supported: Data Protector or Veritas NetBackup. Customer must perform daily and weekly full backups to qualify for the 95% guarantee. Customer must retain their stored data for a minimum of 90 days. Customer data might not be encrypted or precompressed. Only specific data types are applicable for GPG; see the “Terms and Conditions” for details. Note Learn more at http://hpe.com/storage/getprotected http://hpe.com/storage/storeonce

and

||||||||||||||||||||

||||||||||||||||||||

Storage Optimizer HPE Storage Optimizer, shown in Figure 6-22, gives the ability to analyze files based on metadata, so the total amount of storage required for unstructured data—across repositories, data sources, and the enterprise—can be seen, understood, and reduced. Key features include: Analytics-driven: improves data quality and intelligence Connectivity to a wide range or repositories including File Shares, Hadoop, SharePoint, HPE StoreAll, and many others State-of-the-art scalability that is proven at scales of hundreds of terabytes Allows you to stub selected documents, move and delete selected data, and apply complex policies Built-in role-based security, review processes, and full auditing Identifies deduplicate data across different repositories and different storage tiers

Figure 6-22 Storage Optimizer Allows to intelligently identify data currently stored in Tier-1 primary storage infrastructure that can be moved to Tier-2 storage; finds redundant, obsolete, or trivial data to reduce the volume of data stored and applies complex policies and enriched metadata. No longer is storage optimization “blind.” HPE Storage Optimizer combines the power of file analytics and prioritized backup in one cost-effective storage optimization

Technet24

||||||||||||||||||||

||||||||||||||||||||

solution. Customers will get more value from existing data and infrastructure while cutting OPEX (operating expense). By reducing the total volume of data being stored (via better archiving and stubbing), backup times can be faster up to 50% while increasing performance. Note For more information on the HPE Storage Optimizer tool, open the hyperlink. https://www.hpe.com/h20195/V2/GetPDF.aspx/4AA5-9461ENW.pdf

One Config Advanced How to import an Excel file in OCA HPE’s new cloud-based configuration tool, OCA, shown in Figure 6-23, will replace Watson and Sales Builder for Windows to bring major improvements by allowing users to build configurations faster and easier.

||||||||||||||||||||

||||||||||||||||||||

Figure 6-23 One Config Advanced Use OCA for large and complex configuration solutions, and take advantage of its intuitive, web-based interface that captures the customer’s intent documentation within the configuration experience. The tool also provides an effective method to search, clone, breakout, filter, select additional column attributes, and save favorite subconfigurations for reuse.

How to transition from SBW to OCA SBW configurations cannot be loaded directly into OCA. Users are advised to transfer SBW configurations into OCA by using the import Excel option. Transitioning configurations can only take place once the products are available in OCA.

Technet24

||||||||||||||||||||

||||||||||||||||||||

SBW will no longer receive Knowledge Base updates; however, users will still have the ability to export the Excel file that can then be uploaded into OCA. Individual users should export their configurations in an Excel format from SBW Quoter and import and save them in OCA. If SBW documents are used by several users belonging to the same company, then one user should be responsible to migrate the SBW documents into OCA. After this step, the OCA configurations can be shared with other users via the UCID (ID of the configuration) numbers or the OCA local files. A reseller can delegate configurations to a distributor, or the OCA local file can be used to share configurations between companies. These options are available for distributors too. Related to obsolete components, some of them might be retrieved/visible in OCA. If such obsolete products/configurations cannot be managed by OCA, it is recommended to follow the QuickSpecs guides information.

How to import an Excel file in OCA Figure 6-24, Figure 6-25, and Figure 6-26 show the procedure of importing an Excel file into OCA. Note The Export BOM Differences button will not be available once import is completed. Users cannot go back later to see this report without restarting the Import process.

||||||||||||||||||||

||||||||||||||||||||

Figure 6-24 How to import an Excel file in OCA

Figure 6-25 Import BOM to Create OCA Configuration

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 6-26 How to import an Excel file in OCA

Configuring and quoting Nimble Storage The very simple product structure dramatically simplifies configuring and quoting, on every quote: Base array and mandatory support Optionally upgrade/expansion and mandatory support Optionally (some regions only) installation services Products and services are fulfilled by Nimble Storage, an HPE company.

Easy configuration and quoting of Nimble Storage in OCA The Nimble Storage configuration and quoting process in OCA, shown in Figure 6-27 and Figure Figure 6-28, includes following steps: Pick Nimble Storage array from list. Configure available expansions and upgrades. Add mandatory support for the components. Add optional installation services (US and Canada).

||||||||||||||||||||

||||||||||||||||||||

Figure 6-27 Configuring and quoting Nimble Storage

Figure 6-28 Easy configuration and quoting of Nimble Storage in OCA

Additional tools These HPE tools can be used for sizing a storage architecture: HPE Converged Infrastructure Solution Sizer Suite—Solution Sizers from HPE are conveniently available through the HPE Converged Infrastructure Solution Sizer Suite (CISSS). CISSS provides an easy way to install sizers, consolidate the BOMs generated by multiple sizings, access reference architectures, and more.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Protection program—The HPE StoreOnce Get Protected Guarantee Program is an HPE StoreOnce sales enablement program that helps HPE sales and channel partners show customers how to maximize their backup infrastructure and reduce backup capacity needs.

EMC Data Domain analyzer You can use the EMC Data Domain (DD) analyzer to evaluate the “autosupport” file of an installed base EMC DD system. It generates a detailed report about the existing EMC DD environment, including a swap-out proposal and sizing for a StoreOnce solution. To complete the process, follow these steps: Ask your customer to forward their DD autosupport email. Group all files in one zip and attach it to your request email. Submit the request by forwarding it to [email protected]. A few hours after processing, you will receive an email with your assessment report. Whenever possible, HPE recommends analyzing multiple datasets (daily autosupport files). Multiple assessments provide a better view of the customer workload and make it easier to analyze multiple dates with a single request.

||||||||||||||||||||

||||||||||||||||||||

Learning check From where can you download the storage sizing tools?

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Completing the OCA introduction Access the Partner Portal at https://partner.hpe.com/group/upp-emea/oca1 for this activity. Under “Training Material,” select the link to One Config Advanced Videos, and complete the Brainshark E2E training.

||||||||||||||||||||

||||||||||||||||||||

Using TCO and ROI tools The key to presenting a good business case is qualitative and quantitative data about the entire lifecycle cost of a technology project and the impact it will have on business processes. TCO measures cost, and ROI measures benefit. TCO data should feed ROI data, and ROI data should feed the overall business case for business technology decisions.

TCO TCO costs can be difficult to quantify. It is important to ask the customer meaningful questions about why a technology initiative exists and what impact it will have on the business. TCO represents total direct and indirect costs over the entire lifecycle of a hardware or software product. The simple approach to TCO data collection and assessment is a template that requires collecting specific hard and soft data. Hard data is always preferred over soft data, but soft data should be analyzed if it can be monetized (for example, generating a premium for a company’s stock price or enhancing the brand). Examples of soft costs include the cost of downtime, consulting from indirect sources, and costs connected with standardization. Note TCO is a financial estimate intended to help buyers and owners determine the direct and indirect costs of a product or system. It is a management accounting concept that can be used in full cost accounting or even ecological economics where it includes social costs. (March 2018, Wikipedia.org)

ROI ROI can be difficult to calculate. For example, companies developed websites for a variety of reasons in the mid-to-late 1990s. First-generation sites were essentially company marketing displays, but very few transactions took place. What was the ROI for these sites? They did not reduce costs; in fact, they increased them. They did not generate revenue. Companies built them to convince customers, Wall Street analysts,

Technet24

||||||||||||||||||||

||||||||||||||||||||

investors, and others that they understood that the World Wide Web was important. The ROI for this endeavor provided an intangible benefit, meaning the companies might have improved their credibility or reputation, which might have affected downstream profits. Note ROI is the ratio between the net profit and cost of investment resulting from an investment of some resource. A high ROI means the investment's gains compare favorably to its cost. As a performance measure, ROI is used to evaluate the efficiency of an investment or to compare the efficiencies of several different investments. (March 2018, Wikipedia.org) There are many factors to consider while calculating ROI. Research suggests that although many ROI methods are used, the most popular approaches calculate cost reduction, customer satisfaction, productivity improvement, and contributions to profits and earnings. Most business technology executives consider two years to be a reasonable timeline for measuring ROI. A few methods of calculating ROI are: Payback is one simple approach to ROI data collection and assessment. This method calculates the time it takes to offset the IT investment through increased revenues or reduced costs. If the payback period is short and the offsets are great, then the ROI is significant. Payback should be defined by using internal metrics. The payback for some projects will be a year, but others might take three years. Most IT projects should achieve a positive ROI within three years. Another way to determine ROI is based on a simple calculation that starts with the amount of money needed to purchase an IT solution (including TCO and other data). Then, the increased revenue or reduced costs that the investment would generate are projected. If a project costs $1 million but saves $2 million, then the ROI is healthy. Additional ROI methods are based on financial metrics such as economic value analysis (or economic value added), internal rate of return (IRR), net present value (NPV), total economic impact (TEI), rapid economic justification (REJ), information economic (IE), and real options valuation (ROV) among others.

Alinean ROI and TCO analysis Alinean is a well-known ROI and TCO calculator and template designer. The company

||||||||||||||||||||

||||||||||||||||||||

has developed more than 100 ROI sales tools for HPE and several other companies. It created the industry-standard software for chief information officer budgeting, planning, and ROI and TCO benchmarking. Alinean offers a proprietary DB of financial and IT performance information for 20,000 worldwide corporations. It also has proprietary research methodologies to quantify the costs and benefits of IT projects. Alinean software helps users demonstrate ROI, TCO, and the overall value of IT solutions. Note To access the Alinean tools (login required), open the hyperlink: https://www.hpe.com/us/en/solutions/tco-calculators.html Alinean analysis tools, shown in Figure 6-29, use this data to perform calculations and determine the benefits of migrations to HPE solutions: Server, power and cooling, and operating system and DB license costs Operations and administration costs

Figure 6-29 Alinean ROI and TCO analysis Popular Alinean tools you can use in your storage presentations include:

Technet24

||||||||||||||||||||

||||||||||||||||||||

Peer Comparison ROI Analyst Value Finder

HPE Storage Quick ROI Calculator The HPE Storage Quick ROI Calculator, shown in Figure 6-30, can make it very easy to build the business case. The Alinean tool set contains several important HPE templates and calculators.

Figure 6-30 HPE Storage Quick ROI Calculator Alinean tools are available for these products: HPE 3PAR StoreServ HPE StoreOnce Backup HPE StoreVirtual When you are working with decision makers, it is important to provide financial justification. That means presenting a business case that includes the value proposition

||||||||||||||||||||

||||||||||||||||||||

as well as essential financial metrics. The Storage Quick ROI Calculator provides these metrics along with the NPV and IRR. Sales skills combined with Alinean tools and financial metrics will help you develop a compelling business case. In the financial metrics generated by the Storage Quick ROI Calculator, the top half of the chart will show the year-by-year cash flow and the cumulative cash flow. The rows highlighted in orange show the ROI, NPV, IRR, and the payback period. Remember that the output charts within the Alinean tool are based on the customer’s information, which is important to know when developing or proposing an HPE solution.

Alinean customer deliverables The Storage Quick ROI Calculator allows creating of a complete business value report. Using key inputs from customers, a report can be generated that has third-party credibility along with the ability to dig deeper when necessary. PowerPoint presentations, interactive surveys, assessment summaries, and blueprints can be generated using Alinean tools. These customer deliverables include information such as customer data, financial metrics, and competitive analysis. Figure 6-31 shows HPE Converged Infrastructure (CI) Business Value Calculator.

Figure 6-31 HPE Converged Infrastructure Business Value Calculator

Technet24

||||||||||||||||||||

Technical Enablement Tools and Resources From the Partner Portal, a number of Technical Enablement Tools and Resources can be accessed: A quick reference guide provides a brief description of the tool and how to navigate to it. The quick reference guide organizes the tools and resources by stages of the sales cycle. –

Understand—TA, Assessment Tools

–

Position—Product and solution information, presentations, proposal tools

–

Demonstrate—Demo tools and simulators

–

Design and configure—Multiple configurators, sizing tools, capacity advisors

–

Implement and Support—Software Depot

–

Training—TekTalks, MyLearning

||||||||||||||||||||

Resources and tools HPE provides a variety of tools, resources, and other support to help you in your role as a solution architect (SA) generalist or accredited technical professional (ATP). Information available at the enterprise level includes: White papers address key topics and technologies relevant to a particular technology or product. Some white papers are suitable for sharing with customers, but others have restricted information. Note To access HPE white papers and case studies, open the hyperlink https://www.hpe.com/us/en/storage.html#.VfCBCxFVikp

Solution briefs provide key information about a product, service, or supporting activity. The briefs are typically suitable for customer viewing.

||||||||||||||||||||

||||||||||||||||||||

Note To access HPE solution briefs in the HPE Information Library, open the hyperlink. http://h17007.www1.hpe.com/us/en/storage/info-library/index.aspx

Various external web portals compile information from different sources across multiple HPE websites for quick and easy information and tool access. These portals are typically available for corporate HPE customers and channel partners.

HPE Tools website The HPE Tools are an excellent resource when you are strategizing or planning solutions for a storage infrastructure. In addition to providing access to SBW, the HPE Tools site features links to: Price Book Viewer—Allows you to browse product information. It offers a Price Book, which contains product information including short descriptions, prices, and availability data. Optionally, a Price Book can contain discounts and long descriptions if selected. DAR Discount Editor—Is especially valuable to distributor authorized resellers (DARs) to create and modify discount files. The discount file can be used with Price Book Viewer. Note Access the HPE Tools online on the HPE Partner Portal.

HPE Demonstration Portal The HPE Demonstration Portal, shown in Figure 6-32, contains recorded and live demonstrations of HPE technology.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 6-32 HPE Demonstration Portal The HPE Demonstration Portal showcases how HPE technologies lead, innovate, and transform businesses by providing a central location for all demonstrations, webinars, and supporting collateral. Live and prerecorded demos feature HPE hardware, software, services, and partnerships in an exciting multimedia format to show how HPE can help solve your business and IT problems. The HPE Demonstration Portal is an interactive tool that allows you to navigate quickly and easily to find the information most relevant to you. The subject areas show trending, recommended, recent live demos, and recent recorded demos. Choose the area you are interested in, and then select a category within the area to view all available prerecorded and live demonstrations as well as supporting collateral. Prerecorded demonstrations, or demos, provide high-level overview demonstrations of a product or solution, and they are an excellent way to learn more about HPE and partner products or solutions.

||||||||||||||||||||

||||||||||||||||||||

Note Visit the HPE Demonstration Portal at: https://hpedemoportal.ext.hpe.com. Portal contains dozens of live demonstrations, as shown in Figure 6-33, documents, and recorded demos.

Figure 6-33 HPE Demonstration Portal: Live demo

HPE Storage Information Library HPE Enterprise Information Library, shown in Figure 6-34, provides access to HPE Storage information. Users can find there up-to-date information, including installation guides, configuration guides, user guides, references such as release notes, planning manuals, and service and maintenance guides. Note To access this information, visit www.hpe.com/info/docs and click the Storage tab.

Technet24

||||||||||||||||||||

Figure 6-34 HPE Storage Information Library

HPE QuickSpecs The HPE Marketing Document Library allows users to: Access and search the QuickSpecs online, as shown in Figure 6-35. Download an offline QuickSpecs application. Create quick quote for your desired product. Look up individual product list prices. QuickSpecs provide technical information for: –

||||||||||||||||||||

HPE products

||||||||||||||||||||

–

HPE services

–

HPE solutions

||||||||||||||||||||

Figure 6-35 HPE QuickSpecs The HPE Marketing Document Library for QuickSpecs, accessible from your desktop or mobile device, is a convenient central resource providing technical overviews and specifications for HPE hardware and software. It offers full text search, faceted navigation, and search results sorted by most popular documents. Previous versions of QuickSpecs might also be viewed. Note Go to www.hpe.com/info/qs for online access or to download the offline Product Bulletin.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE SAN certification and support The HPE SAN Design Reference Guide, shown in Figure 6-36, provides information on SAN design philosophy and considerations. The guide includes design rules concerning SAN topologies and supported configurations.

Figure 6-36 HPE SAN certification and support In addition to sections on security and management, the guide details HPE best practices regarding planning, upgrading, and migrating SANs. This guide can help you become more familiar with the components of a SAN and can help when designing a SAN for a customer. HPE provides standard configurations ready for implementing, and these designs can be modified by more experienced designers to create a customized solution. The SAN Design Reference Guide: Addresses goals and includes:

||||||||||||||||||||

–

Architectural guidance

–

HPE support matrices

–

Implementation best practices

–

Incorporation of new technologies

||||||||||||||||||||

Includes IP storage implementations such as integrated Small Computer System Interface (iSCSI), NAS/SAN Fusion, FCIP (Fibre Channel over IP), FCoE (Fibre Channel over Ethernet), and DCB (data center bridging). Provides the benefit of HPE engineering while building a scalable, highly available enterprise storage network. Documents HPE Services SAN integration, planning, and support services. This reference document provides information about HPE SAN architecture, including Fibre Channel, iSCSI, FCoE, SAN extension, and hardware interoperability. Storage architects and system administrators can use this document to plan, design, and maintain an HPE SAN. Note To download HPE San Design Reference Guide, go to: https://h20566.www2.hpe.com/hpsc/doc/public/display?docId=c00403562

Single Point of Connectivity Knowledge HPE Single Point of Connectivity Knowledge (SPOCK), shown in Figure 6-37, is the primary portal used to obtain detailed information about supported HPE Storage product configurations. In SPOCK, you will find details of supported configurations (hardware and software) as well as other useful technical documents. Note An HPE Passport account is needed to access SPOCK. To access SPOCK, open the hyperlink: https://h20272.www2.hpe.com/spock/

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 6-37 Single Point of Connectivity Knowledge

||||||||||||||||||||

||||||||||||||||||||

Learning check To obtain hands-on access in the Nimble Live Demo portal, how many days’ lead time do you need to wait?

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Using the SAN Design Reference Guide Instructions: Scenario: You are using a Linux operating system and need to confirm that it will work with the proposed 3PAR 9450 array. Access SPOCK at https://h20272.www2.hpe.com/SPOCK for this activity (HPE Passport required). Check for the compatibility of 3PAR StoreServ 9450 FC with an Ubuntu operating system.

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Proof of Concept As the all-flash startups have entered the storage market, PoCs have become more and more common. The trend began with smaller startups who were able to install a system in hours to demonstrate their ease of use and performance. This strategy spread to other vendors also so that today many of our customers require us to participate in a POC to win their business.

Test guidelines Each test is designed as a standalone test, and as such, the tests should not be run together unless specified in the test sequence. This would not provide a good representation of any product. Each test is designed to highlight a specific flaw in the product. Once tests have been run, ensure that you discuss the results of each test with your customer. If your customer does not wish to run a test, do not push them as this can damage your credibility. Instead, discuss what would have happened if the test had been run. If you are unsure of what a test is designed to highlight, speak to your storage presales consultant. Read through all tests before starting the test sequence.

Resilience Tests Resilience tests are designed to test the availability of the array—for example, what happens during component failure or link failure. These often result in a loss of performance, although in some cases tests might lead to a total failure of the array or even data loss, so ensure that you prepare your customer for the possibility of data loss.

Functionality Tests These tests are designed to show if the technology being reviewed lacks some of the basic features likely to be required by your customer. These should not affect performance or involve data loss so are safe to use on systems holding production data. The customers’ responsibilities include:

||||||||||||||||||||

Provide Power and networking. Provide the success criteria (or confirm the criteria listed in “Decision criteria”). Deliver resources to build test environment and perform performance testing. Provide resources to work with HPE on array configuration. Deliver facilities for testing, evaluation, and collaborating with vendor. Provide input into the PoC planning, evaluation, and documentation. The Vendor responsibilities include: Provide test equipment along with individual expertise to help run the tests. Deliver best practices documentation regarding setup and configuration of the hardware, OS, DB, and applications.

Decision criteria Objectives that must be met to determine success of the PoC: Array performance—Array must be able to successfully complete all the performance tests with a positive result. Array efficiency—Array must be able to use space efficiently with a combination of thin provisioning, data deduplication, and thin snapshots. Demonstrate the ability to present significantly more capacity than the physically installed capacity in the array. The array must be able to successfully demonstrate snapshots, quality of service (QoS) capabilities, and ability to nondisruptively change Redundant Array of Inexpensive Disk (RAID) types. Array resiliency –

Array must be able to ensure data consistency at all times.

–

Array must be able to withstand a complete power outage while maintaining full data integrity.

–

Array must have 1+1 or minimum N+1 resiliency in all major hardware components.

||||||||||||||||||||

HPE PoC App The HPE PoC App, shown in Figure 6-38, is a tool to help run a PoC while providing flexibility and choice to meet the needs of a PoC. Other PoC tools lock you into a

Technet24

||||||||||||||||||||

||||||||||||||||||||

specific workload choice or require you to learn a synthetic workload tool to gain the flexibility needed to meet your PoC goals.

Figure 6-38 HPE PoC App The HPE PoC App tool itself is available through the sales team and cannot be redistributed. The HPE PoC App breaks the PoC into four manageable steps with appropriate automation where it is needed to make the job easier, but not so much that it locks you into someone else’s choices.

Step 1: Prepare the PoC Environment Unfortunately, no tool can remove the hard work of setting up the test environment, but the HPE PoC App offers some tools and the flexibility to reach your goals. The goal of this step is to connect one or more driver systems to a storage array to be tested and configure them to run the test. Elements of this step include provisioning the servers, loading the OS, creating the SAN, and provisioning the storage. Workload and tests

Drivers are servers that will generate the I/O workload for the array to process and then capture the results. The manager is where the user interface to the drivers resides and where the results are consolidated. The manager can share server hardware with a driver or reside on its own server. This

||||||||||||||||||||

||||||||||||||||||||

version of the HPE PoC App uses vdbench software to generate the I/O workloads, so the drivers can be running any OS supported by vdbench. HPE’s recommendation is to use a workload representing the production workload during a PoC. No insight can be gained into the expected behavior of a storage array by running 100% write workload, for example, unless the production workload is also 100% write. The HPE PoC App was designed to offer flexibility in workload choice. The SAN provides the connectivity between the servers and the storage array that will be tested. It is important to configure the SAN to represent the production SAN as closely as possible. If the SAN becomes a bottleneck, the expected level of storage array performance cannot be reached. The most commonly overlooked attribute of the SAN in a PoC environment is lack of sufficient connections. Ensure that the SAN used for the PoC has sufficient throughput to handle the expected workload(s). The storage array being tested must have sufficient capability to represent the production workload. Running PoCs using unusually small configurations (for example, 8 drives) or “as is” equipment (for example, broken controller) will tell you how unusually small or broken equipment will work. This is seldom helpful in understanding the real-world behavior of your production environment. Preparing the PoC environment is often the most labor intensive step, but a very important step. When the environment is configured, specify the servers and LUNs to use for the test in the workload file.

Step 2: Define the Workload The choice of what workload to use is the most important part of the process. If maximum effort is expended completing the other steps to perfection, but the workload choice is not reflective of the production workload, the results will have little value. Select an I/O workload that is representative of how you intend to use the array in production.

Step 3: Execute the Test Executing the test is where automation can help the most. Tests can take a long time to execute, and capturing the results can be tedious without the help of some good tools. The HPE PoC App automates the entire process of executing the tests and capturing the results.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Step 4: Results Analysis and Reporting The HPE PoC App consolidates the data and produces a summary report of the results on completion of the test execution. The results are produced in a single pdf file for ease of review and sharing. Testing should also include Resilience tests. HPE 3PAR StoreServ has been designed to deliver 99.999% or greater availability. The purpose of these tests is to create or simulate failures to validate the redundancy and resiliency of the components of the array such as the Array, Nodes, Disk drives, Disk enclosure, and Ports. It will also validate the mesh-active configuration of the array across all the controller nodes.

VSA design considerations The software-defined primary storage offering, HPE StoreVirtual VSA, is a virtual storage appliance that creates highly available shared storage from direct-attached storage in VMware vSphere® and Microsoft® Hyper-V environments. Its platform flexibility allows you to create a virtual array within and across any x86 server and nondisruptively scale capacity and performance as workload requirements evolve. You will find that regardless of whether your solution is based on vSphere or Hyper-V, most of the overarching design considerations presented are very similar and apply to both platforms. In fact, performance and capacity sizing of the StoreVirtual VSA is the same on the two platforms. Even though the actual implementation of networking differs between those platforms, most of the general guidance apply to both. One of the major advantages of StoreVirtual VSA is its linear scalability for capacity and performance. As a direct result of this distributed and scale-out architecture, it is key that all storage nodes (that is, the StoreVirtual VSA and its storage) provide the same capacity and performance to the storage pool—the StoreVirtual Cluster. The number of StoreVirtual VSAs and their characteristics determines the overall cluster characteristics. Looking more closely at the storage nodes, the requirement to have similar performance and capacity per storage node translates into identical (or at least almost identical) hardware components in the servers running the StoreVirtual VSA. Three core areas that require careful planning are Server platform, Storage options, and Network. Because storage resources are the bottleneck in most environments, StoreVirtual designs

||||||||||||||||||||

typically start out by selecting the right storage options to accommodate capacity and performance needs. The number of disks and RAID controllers on a platform (along with the CPU and memory options) dictates the server model that should host the StoreVirtual VSA and applications running in virtual machines. StoreVirtual uses an Ethernet network as a “storage backplane” and as a way to present storage via iSCSI to servers and virtual machines, so this becomes another key component. You might design and plan to build your storage on StoreVirtual VSA using a four-step approach: Understanding your performance, availability, and capacity requirements –

How much capacity do you need?

–

How much performance do you need?

–

How does your data need to be protected?

Understanding available storage options and configurations –

Which disk technologies can I choose from? Which type is recommended for which workload?

–

What RAID controller technologies can benefit the solution?

–

What are typical RAID configurations and recommended disk layout?

Understanding networking considerations –

How many network adapters should I plan for?

–

What are the network infrastructure requirements?

Understanding your server requirements –

How many resources should I reserve for StoreVirtual VSA?

–

How much CPU and memory do I need for StoreVirtual?

||||||||||||||||||||

When designing a software-defined storage solution based on StoreVirtual VSA, always be sure to understand the performance and capacity requirements of the environment, especially during peak activity. After the storage requirements are understood, a solution can be designed with the right disk drives, server platforms, and network infrastructure.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check Name the two types of tests that are required for a POC.

Write a summary of the key concepts presented in this chapter.

||||||||||||||||||||

||||||||||||||||||||

Summary HPE presales consultants and partners need to learn as much as possible about the customer’s business situation and needs to convey value to each customer. HPE helps partners and customers create the best solution for each IT problem by providing multiple sizing and planning tools. TCO and ROI calculations help present qualitative and quantitative data about the entire lifecycle cost of a technology project. HPE offers a full spectrum of online resources to help you work with customers and implement successful storage solutions.

Technet24

||||||||||||||||||||

||||||||||||||||||||

7

Administering and Monitoring HPE Storage Solutions

||||||||||||||||||||

||||||||||||||||||||

OBJECTIVES After completing this chapter, you should be able to describe the infrastructure monitoring tools to administer and monitor: Hewlett Packard Enterprise (HPE) 3PAR StoreServ systems HPE StoreOnce systems HPE Nimble Storage systems

Technet24

||||||||||||||||||||

||||||||||||||||||||

Prelearning check There are a number of ways to manage a 3PAR StoreServ array. Name four options.

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR manageability There are a number of ways to manage a 3PAR StoreServ array, as shown in Figure 7-1 and Figure Figure 7-2; however, full functionality is provided through the Command Line Interface (CLI) and StoreServ Management Console (SSMC) Graphical User Interface (GUI). 3PAR SSMC—Powerful, complete, and centralized Management Web Application Program Interface (API)—RESTful Interface CLI—Very rich, complete command set. Can be complemented with the 3PAR PowerShell toolkit SMI-S (Storage Management Initiative Specification)—Management from third-party management tools Powerful analytics using InfoSight External Key Manager (ESKM)—HPE Key Manager or SafeNet KeySecure Service Processor (SP)—Physical server or virtual machine (VM) (vSphere or HyperV VM) Call Home, array maintenance, health checks

Figure 7-1 HPE 3PAR manageability

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 7-2 3PAR—Management options

3PAR—Management options HPE 3PAR SSMC is the HPE 3PAR management and reporting console that offers converged management of both file and block on HPE 3PAR StoreServ Storage systems.

HPE 3PAR StoreServ administration HPE 3PAR SSMC, shown in Figure 7-3, offers a modern look and consistent feel and a common interface and language with HPE Management tools such as HPE OneView. Designed to use the latest API and UI (User Interface) technologies, HPE 3PAR SSMC centralizes all HPE 3PAR StoreServ Management under a single pane of glass and offers converged management and reporting for both file and block.

||||||||||||||||||||

||||||||||||||||||||

Figure 7-3 HPE 3PAR StoreServ Management Console Accessible through web-based UI, you can run your operations as HPE 3PAR SSMC tasks with an improved user experience and better responsiveness. All the information you need from HPE 3PAR StoreServ is available at a glance with customizable reporting capabilities, eliminating the need for add-on software tools or diagnostics and troubleshooting. Assess what is happening across the entire data center in seconds via a simple dashboard, and you are just one click away from collecting configuration and health information on any resource. Currently, the SSMC supports up to 32 arrays on one installation. However, this number might vary depending on the number of objects being managed. HPE 3PAR All-inclusive single system software is bundled with all new 8000, 9000, and 20000 HPE 3PAR arrays and is not available for purchase separately. Note HPE Passport account holders can access the current list of supported configurations on the Single Point of Connectivity Knowledge (SPOCK) website. To see the list, open the hyperlink: https://h20272.www2.hpe.com/spock/

HPE 3PAR StoreServ Management Console Overall Enhancements: View alert notifications for all HPE 3PAR StoreServ systems at any time.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Create custom login banner for all users. Customize filters by changing the object’s attributes. Customize the columns displayed in each object’s table views. Improved performance when managing multiple HPE 3PAR StoreServ systems. Dashboard Enhancements: Create your own custom dashboard for one or more systems. Customize panels to view specific systems or key metrics. Quickly switch between consolidated and single system dashboards. New top hosts performance and top volumes panels. View overall counts and status of block and file objects. Add and remove panels quickly and easily. SSMC allows users fast and direct access to context data through the use of links, including: Mega Menu—Links directly to array context areas. Multiple Arrays—Manages up to 32 arrays. Data Aggregation—Aggregates data from all monitored storage arrays into one console.

Data sampling and retention With the HPE 3PAR OS, the SSMC provides information about the on-node data collection period. Data sampling within System Reporter is performed once every five minutes and averaged over time. When a user identifies an object to view, System Reporter retrieves the data for the time requested from the database and displays the information graphically. Results can be viewed at high resolution (five-minute sample), hourly for medium resolution, and daily for low resolution. All sampling is stored in the retention media that System Reporter uses. Early data is displayed at three separate levels including: High—High resolution displays the data collected during each sampling period. The displayed data is the average value over five minutes, where t = time (x) and v = value (y) as sampled at collection time. Medium—Medium resolution uses the same formula as high resolution but will display

||||||||||||||||||||

the aggregate of the data for a one-hour period. Low—Low resolution is the average of all samples taken within a 24-hour period on the array for the object, which is being monitored. The aggregate value is divided by 288, which represents the amount of five-minute intervals for 24 hours.

Using the SSMC The SSMC design focuses on common user tasks. This approach to storage management makes better use of the time IT operations staff has available by allowing them to concentrate on specific areas of the data center. Automation combined with a consumerinspired user experience simplifies basic tasks and everyday processes on HPE 3PAR StoreServ arrays. The design of the SSMC is categorized into three different panels, including: Dashboard—A clear and crisp view of the health, capacity, and performance of the monitored arrays. The dashboard is shown in Figure 7-4. Context areas—Detailed information and submenu items linked to defined context Search—Powerful tool that allows you to filter inquiries to a specific context

Figure 7-4 SSMC dashboard The console is split into four distinct fields: Actions Pull Down—This pull-down menu is used for connection actions such as adding an array for monitoring. –

Connect—Connect a defined array.

–

Disconnect—Disconnect an array from monitoring but do not remove array from list.

||||||||||||||||||||

Technet24

||||||||||||||||||||

–

||||||||||||||||||||

Remove—Remove array from SSMC.

Accept Certificate—Certificates can be exchanged between SSMC and StoreServ. Systems Field—This field identifies the storage systems managed by SSMC and includes a pull-down menu showing the state of an individual array. Session Activity Bar—Located in the upper-right corner is an information field, the bell shows a slide-out message as an actionable event on the array. You can also click on the icon to display the last 100 actions completed. The session icon identifies the current user and a question mark, which is a link to the help page. Main Display—This display is used to identify the current connection and state of the StoreServ array.

||||||||||||||||||||

||||||||||||||||||||

Learning check There are a number of ways to manage a 3PAR StoreServ array. Name four options.

Technet24

||||||||||||||||||||

Activity: HPE Smart SAN for 3PAR video Instructions: Navigate to the following https://www.hpe.com/us/en/storage/3par.html#techSpecs.

website:

Scroll down the page to the “HPE Smart SAN for 3PAR video,” and click View Now, as shown in Figure 7-5.

Figure 7-5 HPE Smart SAN for 3PAR video

HPE 3PAR StoreServ System Reporter HPE 3PAR System Reporter is also fully integrated into SSMC and offers reporting templates, scheduled reports, and threshold alerts. Once an HPE 3PAR StoreServ system is installed, System Reporter automatically collects data on a number of different object data points in the background without the need of any additional setup. With HPE 3PAR SSMC or the HPE 3PAR CLI, a user displays collected data from an array. Although some reports are available to the user without a license, to view all collected data, a license is required. Using the SSMC, graphical reports are available for the following metrics. Historical data –

Performance

–

Histogram

–

Capacity

Real-time data –

||||||||||||||||||||

Performance

HPE 3PAR System Reporter is a feature-rich analytical engine, which helps users interpret collected performance data. With the release of HPE 3PAR OS 3.2.1 MU2,

||||||||||||||||||||

||||||||||||||||||||

System Reporter also incorporates reporting on both block and file HPE 3PAR StoreServ Storage systems. The availability to manage and report on both block and file HPE 3PAR StoreServ Storage systems is available via HPE 3PAR SSMC 2.0. Since HPE 3PAR StoreServ OS 3.1.2, HPE 3PAR System Reporter has moved from an external data structure to the use of an internal data structure designed to retain years’ worth of data created on an internal system volume (srdata). The enhancement in this architecture frees the user from maintaining a separate system and database for monitoring HPE 3PAR StoreServ Systems and allows the system administrator for quick access to data. Collected data can be viewed by creating reports via templates, which are available under the System Reporter section using SSMC; this same section also allows a user to configure real-time performance reporting, which graphs current loads on the array. A new report type called “System Report—Performance Summary” has been added, which provides a complete set of performance charts and configuration information for a system. Reports can be opened (popped out) into a separate window. This makes it easier to see multiple reports at a time. Reports can be toggled between one column and two column views. This allows more graphs to be seen at a time. The x-axis is now at the top of the report, so it is more visible and easier to zoom in and out. Each graph also includes an x-axis label. Tool tips for graphs in the same report are visible at the same time and are synchronized. Lines in graphs are easier to read because of improvements to the line series pattern, consistency in the legend, and tool tips. Real-time reports can be exported to PDF and CSV (comma-separated values) files. “Compare with” reports include multiple systems when exported to PDF and CSV files. SMTP Server Authorization can be enabled for scheduled reports and threshold alerts. Added support for file persona ports to the Host Port Performance template.

System Reporter reports You can access reports generated by System Reporter, as shown in Figure 7-6, by using the Mega Menu and selecting Reports. If you have not previously created any reports, the screen is blank. In the left pane, click +Create report or from the Actions dropTechnet24

||||||||||||||||||||

||||||||||||||||||||

down menu located on the upper-right side of the context pane, click Create.

Figure 7-6 System Reporter reports Data types

Historical data: Charts that collect stored data for the defined object. Each definition of the charted historical object contains a default period for viewing. Each stored historical object can be modified to chart a specific time to suit the user’s needs. Retention time and sampling periods are covered completely in the section titled Database retention. Real-time data: Charts that collect data, which has been collected on an object in the last five seconds. The charted data is a continuous plotting graph of object data. Data charted can be the default object data or user-defined object data.

||||||||||||||||||||

||||||||||||||||||||

Customization

Each object selected for reporting uses a default setting for the particular object that is being charted. The default values for each charted object will differ from object to object. Customization allows the user to select the variables in which each chart is generated. Best practice

Always modify Report name to include a unique identifier for each chart; this customizes the report and allows the user to select the Report template more than once.

Report creation When creating a report, users can select from different report templates. The templates include these categories: Controller Node Host Port Volume Physical Drive Enclosure Port Remote Copy Each of these categories presents several options. When you click +Create report, a pop-up menu appears with all the templates available. To view a template, highlight one (for example, “Exported Volumes—Performance Statistics”) and click the Select button. Example of a report is shown in Figure 7-7.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 7-7 Report in SSMC For each template, a pop-up menu appears with all the options available on selection. The menu includes four distinct areas: General Time reporting and sampling Object selection Chart options Note For a full description of available fields, refer to the HPE 3PAR SSMC administrator guide by opening the hyperlink. http://h20566.www2.hpe.com/hpsc/doc/public/display?docId=emr_nac04777556

Real-time reporting, searching, and filtering After reports are created, they remain on SSMC until you remove them. You can find a particular report using the search function. For example, for a small array and 10 hosts

||||||||||||||||||||

||||||||||||||||||||

attached using a block structure, you can generate 245 individual reports. However, with the search function, you can filter out a subset of reports and choose from the reports listed in the filter; the more filters you include in the search, the better filtering capability.

Templates for real-time reporting SSMC includes templates for real-time reports. These graphs continually update when you are reviewing the report. Every time you go to a different report or screen and return to the initial report, it plots statistics from the latest data point. The previous data points are not shown by default. If you want to see the previous data points, the action called “Get all real-time samples” plots all the data being held in the SSMC back end. SSMC retains 1,000 samples at the back end for each real-time report. There are several options under the Actions Pull Down menu to assist you in reading reports. These options include: Zoom—Allows you to zoom into a time frame during the reporting period Reset Zoom—Starts plotting again from the latest data point Get all real-time samples—Accesses all samples (maximum 1,000) from the back end cache and plot; useful when using Zoom or moving reports When creating a report and choosing a template, rename the template with a unique name. With this extra step, you can reuse the template again. If you do not do this and try to create a report with the same template name, System Reporter displays an error indicating the report already exists. Real-time reporting is shown in Figure 7-8.

Figure 7-8 Real-time reporting, searching, and filtering

Technet24

||||||||||||||||||||

||||||||||||||||||||

Threshold alerts System Reporter enables the storage administrator to set threshold alerts about the performance of system components from within the SSMC. Threshold alerts are predefined templates that might have set values assigned to each alert. Some threshold alert definitions might be shell alerts, which means a template has been defined but no default metric is associated with that alert. It is the user’s responsibility to set these thresholds. Example of a threshold alert is shown in Figure 7-9.

Figure 7-9 Threshold alerts Threshold alerts are designed in categories. Each category contains a number of alert templates. The most common categories include: Controller node Exported volumes Physical drives Port thresholds Other (node link, Quality of Service [QoS], Remote Copy, and Remote Copy volume alerts) Similar to the process of creating reports within System Reporter, the threshold alert templates can be customized. Each of the templates contains these categories: General—Customization in the type of resolution and severity level

||||||||||||||||||||

Objects—Customization in selection of objects to be monitored for alerting –

All

–

Filter by objects

–

Filter by rules

||||||||||||||||||||

Metrics—Object threshold which is monitored and the values of the monitored object If an alert is not part of the predefined alert templates contained within the SSMC, the user can create a custom event alert using the HPE 3PAR StoreServ CLI. After the custom alert is created, it can be viewed in the SSMC as part of the threshold alerts. An alert threshold notification can be viewed from several sources. The SSMC dashboard will display performance status by using two colors: Green—No alerts have been recorded. Yellow—Some alert criteria have been met. If some alert criteria have been met, System Reporter displays the performance alert on the dashboard. Clicking on the performance graph links the user to the Threshold Alerts page within the SSMC. You can also monitor the Threshold Alert page from the main menu. In the main menu, an informational banner displays the alert criterion. The user can get more details about the alert by clicking on the notification.

Excel add-in for SSMC The HPE 3PAR Excel add-in allows users to extract and report data from the System Reporter database (.srdata) in Microsoft Excel. The add-in extracts data using the SSMC server and provides an accessible way for users to create reports including performance data. To install the HPE 3PAR Excel add-in, follow these steps: Download the .iso file for the HPE 3PAR Excel add-in from HPE Software Depot. Exit Microsoft Excel and close the program. On the client system, run setup.exe and follow the instructions. The add-in installs the binaries to the default path, which is C:\ProgramFiles\HewlettPackard\HP3PARSRExcelAddin or to C:\Program Files (x86)\HewlettPackard\HP3PARSRExcelAddin.

Technet24

||||||||||||||||||||

||||||||||||||||||||

To use the HPE 3PAR Excel add-in, shown in Figure 7-10, follow these steps: Launch Microsoft Excel. Select the ADD-INS tab, and then click HPE 3PAR Excel Add-in. Enter the SSMC server name and port, and then enter the username and password (HPE 3PAR storage system credentials). Click Connect to SSMC. TIP When you generate performance data, scroll to the top left of the Excel spreadsheet to view the CSV data.

Figure 7-10 Excel add-in for SSMC

HPE 3PAR Service Processor The HPE 3PAR Service Console (SC) is the new interface for the physical or virtual SP, which collects data from a managed HPE 3PAR StoreServ Storage system in

||||||||||||||||||||

||||||||||||||||||||

predefined intervals as well as an on-demand basis and sends the data to HPE Remote Support, if configured. The SP is available in physical and virtual versions for the SP 5.0 release. The SP is designed to provide remote error detection and reporting and to support diagnostic and maintenance activities involving the storage systems. The SP is composed of a Linux operating system and the SP software, and it exists as a single undivided entity. The physical SP is a hardware device mounted in the system rack. If the customer chooses a physical SP, each storage system installed at the operating site includes a physical SP installed in the same cabinet as the controller nodes. A physical SP uses two physical network connections; one (eth0) requires a connection from the customer network to communicate with the storage system; the other (eth1) is for maintenance purposes only and is not connected to the customer network. The Virtual SP (VSP) software is provided in an Open Virtual Format (OVF) for VMware vSphere Hypervisor and self-extractable Virtual Hard Disk (VHD) package for Microsoft Hyper-V. The VSP is tested and supported on Microsoft Hyper-V (Windows Server 2012/2012 R2/2016) and the VMware vSphere hypervisor (VMware ESXi 5.5/6.0/6.5). The VSP has no physical connections. It runs on a customer-owned, customer-defined server and communicates with a 3PAR StoreServ Storage system over its own Ethernet connections.

Service Console

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 7-11 Service Console The 3PAR Service Console (SC) also allows service functions to be performed by a company admin, HPE support, or an authorized service provider. SC replaces Service Processor Onsite Customer Care (SPOCC), the GUI for SP versions 4.4 and earlier. SC functionality is similar to SPOCC. From Actions, you can check updates for your System O/S or the SP. Main menu & banner: Clicking the menu area opens the main menu of screens. Selecting a screen from the menu displays a summary of configuration settings for the selected screen and provides Actions menus. Depending on the screen selected, system resources such as CPU usage, throughput, physical memory, and so on might be displayed. Screens: The screens area displays tabular and graphical information and provides action dialogs for managing storage systems. Detail pane: Displays the detailed information that can be seen and allows you to select further views. List pane: When you select an item in the List pane, additional information about it is displayed in the Detail pane. When an item is selected, you can perform actions on it. Many lists can be sorted and filtered and include multiple views that you can select. Actions: The Actions menu allows you to perform actions on one or more resources that

||||||||||||||||||||

||||||||||||||||||||

you have selected, in the list pane. If you do not have permission to perform an action, the action is not displayed in the menu. Also, some actions might not be displayed because of system configurations, user roles, or properties of the selected resource. Remote benefits of the SP include: Remote diagnostics—Key diagnostic information maintained centrally on a historical basis. System health statistics, configuration data, performance data, and system events can be transferred frequently and maintained centrally. As a result, proactive fault detection and analysis are maximized, and manual intervention is minimized. Remote serviceability—Provides fast predictive response and remediation. HPE 3PAR technical support delivers rapid responses with 24x7 remote monitoring and analysis to identify issues and proactively communicate them back to the customer. As an integrated support model, HPE 3PAR technical support can remotely connect to a customer’s HPE 3PAR StoreServ system through a secure IP connection to solve issues quickly and reduce onsite visits. Remote online software upgrade—Upgrade software with no application disruption. Changes to the HPE 3PAR OS software are released for new functionality, maintenance updates, and software patches. The ability to apply these updates can be serviced as an online upgrade, where the arrays’ capability to process customer data does not need to be disrupted during the software upgrade. Additional features of the HPE 3PAR SP include: Collects periodic data from HPE 3PAR StoreServ (alerts, configuration, events, performance, status, and so on). Performs hourly health checks on the HPE 3PAR StoreServ system. Provides remote support capabilities over Ethernet. Serves as a local maintenance terminal for onsite support. Serves as a remote maintenance terminal for remote support. Contains guided maintenance scripts used for guided parts replacement activities.

VSP infrastructure planning and deployment best practices The SP can be deployed in two modes: SP mode or Secure Network mode (which is recommended). SP mode enables communication between an SP (version 4.3 and below) and the HPE 3PAR connection portal server located at HPE. Possible Technet24

||||||||||||||||||||

||||||||||||||||||||

configuration is shown in Figure 7-12.

Figure 7-12 VSP infrastructure planning and deployment best practices In Secure Network mode, the SP communicates with the HPE 3PAR Collector Server using the HPE 3PAR Secure Service Agent (SSA). The Collector Server provides software updates, access to service tool applications, and access to resources such as the HPE Support Center. Rather than using a connection portal to connect to HPE 3PAR Central, an SP in Secure Network mode requires SSA for the connection. The SSA facilitates communication between the SP and Collector Server. Communications are performed using HTTPS. Secure Network mode is the preferred method of transfer because it is secure and offers the most access control options. This mode is easy for users to implement in the firewall. When using HPE 3PAR Virtual Service Processor (VSP) in Secure Network mode, use Port 443 for the bidirectional communication. When deploying the HPE 3PAR VSP, using Secure Network mode is recommended.

Local notification service The SP local notification features enable users to request notifications for important storage system events and alerts on a subscription basis. Notifications are sent through email to all subscribers, with each subscriber specifying up to three email addresses. When Real-time Alert Processing (RAP) forwarding is enabled, copies of all notification messages sent to subscribers are automatically forwarded to HPE 3PAR Central as well. There are two types of local notification messages that users might receive: Standard notification messages—Standard notification is a text-based email message

||||||||||||||||||||

that alerts users to an important event or alert generated by a storage system. Grouped, low-urgency notification messages—A grouped, low-urgency notification is a text-based email message that informs users of noncritical events generated by a storage system. –

||||||||||||||||||||

Low-urgency notification messages are informational and do not typically require any corrective action. When a situation or event reported in a low-urgency notification message becomes urgent, a standard notification message is issued to alert subscribers.

System support information It is critical that the VSP contains correct system support information. The installation site information and customer contact information are critical to support delivery. Having the latest information will help ensure that service calls are completed without any issues because of incorrect or outdated contact information.

Remote support All HPE 3PAR customers are required to run SP. This enables HPE to provide customers with the best possible remote support for their HPE 3PAR storage systems, including: Timely remote service Remote online software updates Accelerated troubleshooting and issue resolution Remote support sends diagnostic information (for example, system health statistics, configuration data, performance data, and system events) to HPE 3PAR Central. These diagnostics are required for HPE to perform fault detection and analysis on HPE 3PAR StoreServ systems, which helps maximize storage availability. All remote communications are encrypted and transferred securely to HPE 3PAR Central, and no customer application data is ever transferred. No other business information is collected, and the data is managed according to the HPE Data Privacy policy. Remote support can be configured during the initial setup of the HPE 3PAR StoreServ system using the HPE 3PAR SmartStart process in the SP setup wizard. You can validate and test the connectivity using SPOCC. HPE highly recommends having remote support

Technet24

||||||||||||||||||||

||||||||||||||||||||

deployed and configured for your system.

VSP networking configuration HPE recommends that the VSP be configured with a static IP address. Providing network redundancy to the virtual network cards of the VSP enables availability in case of different failure scenarios. Users should consider using network card teaming in the hypervisor when supported. They can also consider using multiple network cards and physical network switches to avoid single point of failure on the hypervisor, which might affect the VSP network connectivity. Example of VSP networking configuration is shown in Figure 7-13.

Figure 7-13 VSP networking configuration Each VSP manages one HPE 3PAR StoreServ system. Users should consider separating each VSP from the HPE 3PAR StoreServ system that it is managing. This would help ensure that the VSP is available to troubleshoot any issues with the array. Users could deploy the VSP on the local disk of the hypervisor or put it on another array that the VSP is not managing. Putting the VSP on shared storage allows users to leverage hypervisor high-availability options or mobility features. VMotion is the vSphere technology used to move VMs across datastores. If VSP runs from an internal, local drive, you can migrate it with the latest vSphere enhancements. HPE recommends running it on different shared storage. It is also recommended that thin provisioning be used when deploying the disk for the VSP. It will reduce the amount of disk space that the VSP consumes.

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check The Excel add-in for SSMC supports Excel 2013 and 2016. True False

||||||||||||||||||||

||||||||||||||||||||

Activity: Prerequisite for setting up the Service Processor Instructions: Use the “HPE 3PAR Service Processor Software 5.0.x User Guide” at: https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-a00006416en_us for this activity. What does the PDF say about the prerequisite login support to update the operating system though the Service Console?

Technet24

||||||||||||||||||||

||||||||||||||||||||

Managing HPE StoreOnce systems Managing and monitoring the StoreOnce from the management software can be completed through the Web Management interface GUI or the CLI. GUI is shown in Figure 7-14.

Figure 7-14 Managing HPE StoreOnce systems Two roles define the permissions associated with admin and user. Administrator: This account allows authorized users to create and edit management and StoreOnce functions via the GUI and CLI. The default login and password are Admin and admin, respectively. The permissions role of the administrator is admin. You cannot delete this account, but you can and should modify its password. Operator: This account limits access to the GUI and CLI to monitoring and viewing. The default login and password are Operator and operator, respectively. The permissions role of the operator is user. The StoreOnce CLI provides an alternative way of administering and monitoring the system. Some tasks are only accessible from the StoreOnce CLI, StoreOnce CLI

||||||||||||||||||||

||||||||||||||||||||

commands require an SSH client application. The HPE StoreOnce page is the initial page at logon. It provides important information about system status and processes that are running. It also provides access to individual StoreOnce functions, which are available as subtopics (click + to display them in the Navigator). System Status (top left side panel): Displays the last time the system was updated and a count of event status categories. Navigator (left side panel): Displays available GUI topics and enables navigation to selected topic pages. Topic page (main window): Initially, the main window displays the top-level StoreOnce topic page. Select another topic from the Navigator to display the page for that selection. Adding a new user: From Device Configuration, (1) select User Management (2) and Add User (3) to create a new user. In the Add New Local User dialog, create a user with the following data: 1. User Type: Local 2. User Name: xxxx 3. User Role: admin 4. Password: xxxxxx 5. Click OK when done.

Reporting Central—Reports The Reporting Central page allows you to quickly view a snapshot of the status and capacity usage of registered appliances. Reporting Central, shown in Figure 7-15, allows you to view the status, throughput, and storage details of local and remote StoreOnce appliances. You can create groups of either appliances or virtual devices and then run reports that pull data from these groups. You can also perform system resource monitoring.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 7-15 Reporting Central—Reports Register any appliance you want to pull reports for: Configure groups containing registered appliances and their devices. Run reports using the configured groups of appliances and virtual devices. Perform resource monitoring. Resource monitoring includes CPU/Memory, Disk Utilization, and Network Utilization.

Firmware updates Select the Firmware page to view the firmware versions for hardware components and to update firmware versions if required. The page, shown in Figure 7-16, contains the following tabs: Server: Contains hardware components for the server. Storage: Contains hardware components for all storage elements.

||||||||||||||||||||

||||||||||||||||||||

Figure 7-16 Firmware updates The tabs are empty when the page is first displayed, and no information will be present until a scan has been initiated.

Configuration save Operation can be done at any time using the following process: Start a putty session to the StoreOnce unit and login with the default Admin credentials. Type the command config save devices , as shown in Figure 7-17.

Figure 7-17 CLI examples Configuration information is stored in the config directory of your system. Type the command config show list devices (or if you are already at the config prompt type the command show list devices . StoreOnce appliances enable creation of a backup of device configuration settings in the form of a file that can be retrieved from the appliance and stored on the user’s PC for possible use in the future. The configuration backup file might be used to quickly

Technet24

||||||||||||||||||||

||||||||||||||||||||

configure a replacement or repaired StoreOnce appliance in the unlikely event that the appliance fails, and all data and configuration details are lost. For example, a failure of the appliance operating system has resulted in the need to re-image the appliance and configuration details and data are lost. The configuration save and restore process is managed via the StoreOnce CLI, and an SFTP (SSH File Transfer Protocol) client is required to download or upload the saved configuration file. The configuration save and restore process does not include encryption key store data. If using this feature, separate StoreOnce CLI commands should be used to back up the encryption key store and store it safely in an offline location. Use the original key store to access encrypted key store data after a restore.

Confidence Checker Start a putty session to the StoreOnce unit and login with the default Admin account credentials. Enter the command system confidencechecker runtest all , as shown in Figure 7-19, or specify a different parameter to define the type of test to perform: All—runs all of the available tests. internalNetwork—tests connectivity of the hardware components and the switch configuration (StoreOnce 6500 and 6600 Systems only). externalNetwork—tests connectivity of the external network including AD (Active Directory) and IPSec. fileSystem—tests the overall health of the file system and the load balancing. Sas—tests SAS (serial attached SCSI) connectivity (StoreOnce 4900, 6500, and 6600 Systems only).

||||||||||||||||||||

||||||||||||||||||||

Figure 7-19 HPE RMC-V configuration in vSphere web client Runs an appliance “self test” and provides a report listing any faults found. The test result will be Passed, Passed with Warnings, or Failed with additional information is provided where appropriate. The example in Figure 7-18 runs a self-test on the appliance, only test is visible, filesystem check, with the result of PASSED.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 7-18 Confidence Checker The HPE StoreOnce 6000 series of modular systems features HPE inline deduplication and network efficient technology and offers extended capacity for Enterprise-class storage. If displayed, click on a service set to display status information for the services running on that set. Click on any column heading to change the sort order or select which columns are displayed. A single-couplet configuration has two Service Sets. This is the minimum configuration. Additional couplets increase Service Sets, two for each couplet, up to a maximum of eight.

HPE StoreOnce RMC for VMware software (RMC-V) HPE StoreOnce Recovery Manager Central (RMC) software integrates HPE 3PAR StoreServ primary storage with HPE StoreOnce systems. This provides applicationmanaged data protection that speeds and simplifies data protection for virtualization data on an HPE 3PAR StoreServ array, bypassing traditional backup server-based processes. The HPE StoreOnce RMC software for VMware is a separate, individually licensed

||||||||||||||||||||

||||||||||||||||||||

product that customers can use with HPE OneView for VMware vCenter. StoreOnce RMC for VMware software provides a converged availability and flat backup service for VMware environments that augments traditional backup approaches. Example of RMC-V configuration in vCenter is shown in Figure 7-19. StoreOnce RMC for VMware is an array-based, online, VM snapshot and recovery solution that gives superior control over data protection and recovery in VMware vSphere environments. This includes granular, rapid online recovery of files. By leveraging HPE 3PAR StoreServ Virtual Copy software, this software enables VMware administrators to nondisruptively create hundreds of space-efficient, applicationconsistent, point-in-time snapshots to protect and recover VMDKs (Virtual Machine Disks), Virtual Machine File Systems (VMFS), individual VMs, or individual files. Along with the HPE 3PAR Management Plug-In for VMware vCenter, StoreOnce RMC for VMware provides administrators access to all of this functionality from within the familiar and easy-to-use VMware vCenter Server virtualization management console. StoreOnce RMC for VMware provides: Application-consistent VM snapshots for rapid online recovery Support for provisioning, volume, drive type, and remote copy Visibility into underlying storage resources to see how each VM is mapped Highly granular and flexible recovery point objectives (RPOs) from the VMware vCenter console Support for Microsoft SQL Server and VMware environments The Express Protect feature offers data protection by facilitating Express Protect of snapshots from HPE 3PAR StoreServ to HPE StoreOnce. Express Protects on HPE StoreOnce are self-contained volumes, deduplicated to save space, and can be used to recover back to the original or a different HPE 3PAR StoreServ Storage system based on the options chosen. Administrators can access all the functionality from within the VMware vCenter Web Client. The HPE StoreOnce RMC Base license is required to take crash-consistent snapshots of any application. This license also enables you to use the RMC REST APIs that you can leverage to create your own application-consistent plug-ins. This license is supported across all the HPE 3PAR StoreServ Storage models. The HPE StoreOnce RMC with VMware license is required for taking VMware application-consistent snapshots. There are array-specific variants of this license. The RMC base license is not required if the application-consistent license is available, since the base functionality is included within this license.

Technet24

||||||||||||||||||||

||||||||||||||||||||

If you are currently using RMV (HPE 3PAR Recovery Manager for VMware vSphere) and have a valid license, you do not need to purchase any of the new RMC licenses mentioned above. Your existing RMV license is automatically detected and used with RMC-V. The HPE 3PAR Virtual Lock license is required for setting retention policies in Recovery Set creation. Types of licenses available for RMC are shown in Figure 7-20.

Figure 7-20 License models supported by RMC

Installing Recovery Manager Central You can install RMC, components are shown in Figure 7-21, on VMware vSphere ESXi or Microsoft Hyper-V hypervisor hosts.

||||||||||||||||||||

||||||||||||||||||||

Figure 7-21 Recovery Manager Central components The RMC installation wizard is used to install RMC on VMware vSphere ESXi. The wizard guides administrators through a set of screens with custom fields. RMC can be installed on Microsoft Hyper-V using the PowerShell Installer script.

Connectivity considerations The connectivity between HPE 3PAR StoreServ and RMC appliance for data traffic is either over: FC (Fibre Channel) iSCSI (integrated Small Computer System Interface) Note For iSCSI, the current release of RMC does not support HPE 3PAR StoreServ ports that are configured with a VLAN (virtual local area network) tag.

Technet24

||||||||||||||||||||

||||||||||||||||||||

The connectivity between HPE StoreVirtual and RMC appliance for data traffic is over iSCSI. The connectivity between HPE StoreOnce and RMC appliance is over: CoFC (Catalyst Over Fibre Channel) CoEthernet (Catalyst Over Ethernet) The connectivity between RMC, HPE 3PAR StoreServ, and HPE StoreOnce for management traffic is over IP. Note The first CoFC backup performed on an RMC instance sets the value of Number of Devices per Initiator Port to 64. In case you are modifying the value, you must manually reboot RMC for the modifications to take effect. If the required number of paths is not available and if you are manually changing this value, the changes might impact backup operations. If you are configuring additional FC ports for HPE StoreOnce, you have to configure Number of Devices per Initiator Port in the HPE StoreOnce Management Console.

RMC—Fibre Channel connectivity RMC appliance supports FC multipathing, as shown in Figure 7-22. Configuring multiple (redundant) FC paths to the Storage Systems ensure effective path failover.

Figure 7-22 Fibre Channel connectivity For VMware, FC connection to RMC is possible only through VMware VMDirectPath.

||||||||||||||||||||

||||||||||||||||||||

FC multipathing enhances: Backup and restore throughput and hence improves RPO and Recovery Time Objective (RTO). Fault tolerance during backup or restore operations, due to any path failure. HPE recommends that you configure a dedicated path to each storage system port and backup system port for faster backup window. Follow the recommended zoning best practices to access the storage and backup systems. Note A maximum of two paths are supported from RMC to HPE 3PAR StoreServ Storage system.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check What are the two tabs in StoreOnce Manager GUI for updating firmware?

||||||||||||||||||||

||||||||||||||||||||

Nimble Storage system Initial setup for administration is completed through the Nimble Setup Manager (NSM), shown in Figure 7-23. NSM is a Windows Utility that uses automatic array discovery. NSM will list the array names, their model numbers, and NimbleOS version. The array must be on the same subnet (or directly attached). Setup can be and can also be completed with the CLI through the serial port.

Figure 7-23 Nimble Storage system At this point, a variety of basic parameters can be configured for the array including: Assigning a new name and group configure a management IP address/subnet mask/gateway Setting up a user password.

Technet24

||||||||||||||||||||

||||||||||||||||||||

NimbleOS home page The NimbleOS home page, shown in Figure 7-24, includes Manage, Hardware, Monitor, Events, Administration, and Help.

Figure 7-24 NimbleOS home page Each dashboard displays various sections: Performance, Space, Protection, hardware, Alarms, and Events.

Events menu The Events menu, shown in Figure 7-25, provides detailed information regarding events that have occurred on the array.

||||||||||||||||||||

||||||||||||||||||||

Figure 7-25 Events menu Using the filter, you can quickly drill down to critical events that need attention for the managed Arrays.

Manage menu Manage menu, shown in Figure 7-26, includes Data storage, Data protection, Data access, and Performance policies.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 7-26 Manage menu Manage is more for daily task and Administration is for more seldom tasks. This is the menu where storage is provisioned, replication configured, and access is granted. All volumes by default are configured from a single large pool and are thinly provisioned. Access to volumes is also done here through the concept of initiator groups that contain WWPN (World Wide Port Name) or iSCSI initiator names. The system tries to make it easy by importing the alias names from the FC switch if they have been configured.

Monitor menu Selecting the monitor menu allows administrators to view various criteria including Capacity usage and savings Performance IOPs, latency and bandwidth, Interface utilization, Replication, Connections, and Audit log information. Various monitor screens are shown in Figure 7-27, Figure 7-28, and Figure 7-29.

||||||||||||||||||||

||||||||||||||||||||

Figure 7-27 Monitor menu

Figure 7-28 Monitor menu/Performance

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 7-29 Monitor menu/Space

Administration menu Software downloads and update can be performed in the Administration menu, along with space defaults, VMware integration with vCenter registration, and VVols (VMware Virtual Volumes). Administration menu is shown in Figure 7-30.

Figure 7-30 Administration menu From Security tab, administrator can set up a new User, as shown in Figure 7-31.

||||||||||||||||||||

||||||||||||||||||||

Figure 7-31 Administration menu User roles available: Administrator (Default “admin” user) Complete access and security capabilities—no limits. PowerUser—Same as Administrator but without the privilege to modify security settings. Operator—Same as PowerUser but without the privilege to delete data and operations that are rarely performed. Guest—Read-Only privilege except for security settings.

How does InfoSight help customers? Technet24

||||||||||||||||||||

||||||||||||||||||||

InfoSight provides nonstop availability with InfoSight Predictive Analytics. Advantages, shown in Figure 7-32, include: Prevent issues and avoid downtime Cross-stack rapid root cause analysis Predict future needs and simplify planning

Figure 7-32 InfoSight advantages The sensors capture health data across installed customers’ base. These sensors are sending us information not only from the storage but also from the other infrastructure components. Deep Data analytics correlate this data with health checks and learning from entire installed base. Nimble uses this information within their Support organization. Nimble also makes the information available in the form of a Software as a Service (SaaS) portal that their customers can access through a browser to monitor any system deployed anywhere around the globe. This health data, like when you take blood pressure or heart rate information, is used to determine steps to improve performance and maintain uptime of the infrastructure stack. The recommendations include prescriptive guidance on fixing configuration and

||||||||||||||||||||

||||||||||||||||||||

interoperability issues in VMs, networking, and so forth. InfoSight also predicts future needs and helps with capacity planning. For example, if you have some number of workloads running on an array, and you want to bring on a new database, InfoSight can model whether there is enough performance headroom or whether you will need to expand the existing system. So what are the main benefits that customers see from InfoSight? First, if you look at all of Support cases, in 9 out of 10 cases, Nimble proactively contacts customers to prevent something that could go wrong. This is a game-changing experience for customers. Second, when a customer does call Nimble, HPE support already has all of the information needed to rapidly resolve problems. No need to gather logs and diagnostic data for hours and even days before issues can be resolved. Nimble does not use any first- or second-level Support specialists. InfoSight handles the work of first- and second-level Support specialists. When a customer talks with Support, they are always talking with a third-level Support engineer. Additionally, the average hold time to speak with a specialist is less than a minute.

Identifying where the problems exist If it was just about storage, things might be a lot simpler. But it is not, it is more complicated, in fact less than one half of cases are directly related to storage. More than half of issues that can contribute to the App-Data-Gap lie outside the storage component, as shown in Figure 7-33.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 7-33 Identifying where the problems exist The fact is, that with the complexity of today’s data center—where teams support applications, compute platforms, networking, storage, and converged infrastructure— enterprises can no longer rely solely on their own individual teams to gather all of the complex operational data themselves and turn it into useful actions. And that is where InfoSight helps.

||||||||||||||||||||

||||||||||||||||||||

Learning check The initial Nimble Array setup must be completed in the CLI. True False

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE StoreVirtual storage systems

Figure 7-34 HPE StoreVirtual Management Console The StoreVirtual Management Console (SVMC) software provides contemporary browser-based interfaces for monitoring HPE StoreVirtual storage. Using the HPE StoreVirtual 3200 System menu, administrators can monitor the overall of the status of the array. Selecting a Storage Pools Capacity doughnut provides details capacity information for configured storage pools.

Dashboard view A dashboard view shows the availability of any outstanding patches or system software version updates, as shown in Figure 7-35. These patches and updates are issued to fix problems or introduce new features. What is more; HPE recommends that they are installed in a timely manner.

||||||||||||||||||||

||||||||||||||||||||

Figure 7-35 Dashboard view

Centralized Management Console main window For StoreVirtual VSA, the Centralized Management Console (CMC) is being used for managing the nodes, as shown in Figure 7-36. Features include: Available Systems—VSA Storage Systems that are powered on but not in a Management Group. Management Groups—Groups of Storage Systems within which one or more modules are designated as Managers. Everything having to do with StoreVirtual VSA storage area network (SAN) Storage is contained within a Management Group. Clusters are subgroupings of Storage Systems within a Management Group. However, a Cluster can be as small as a single Storage System. Volumes—Data storage areas created on Clusters. Volume icon indicates Network RAID level. Snapshots—Read-only copies of Volumes created at specific points in time. Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 7-36 Centralized Management Console main window

||||||||||||||||||||

||||||||||||||||||||

HPE OneView HPE OneView provides converged management capabilities that reduce infrastructure complexity through automation. This modern management architecture is designed to accelerate IT operations for managing servers, storage, and network resources. HPE OneView is purpose built to support scenarios such as: Deploying bare-metal servers Deploying hypervisor clusters from bare metal Performing ongoing hardware maintenance Responding to alerts and outages HPE OneView uses templates that automate the configuration and propagation of server, storage, and networking profiles. Its underlying infrastructure components—including networking, power management, and servers—support virtualization, cloud computing, big data, and mixed computing environments. This converged, software-defined, and automated platform reduces operating expenses (OPEX), improves agility, and frees up resources for new business initiatives. HPE OneView can manage c-class solutions, Synergy solutions including compute modules, and many HPE rack-based servers. HPE OneView also provides variety of storage management tasks. Figure 7-37 shows a StoreServ system managed by HPE OneView.

Figure 7-37 HPE 3PAR StoreServ Management in HPE OneView

HPE OneView storage management Technet24

||||||||||||||||||||

HPE OneView allows administrators to: Automate HPE 3PAR StoreServ volume creation and SAN zoning. Attach the storage volumes to server profiles. For example, a 32-server cluster can be automatically created in hours rather than manually configured in days. After deployment, storage and servers are monitored in HPE OneView, and the storage topology is visible in Map View. Automated, policy-driven provisioning of storage resources is fully integrated with server profiles. You can use SAN managers to bring their managed SANs under HPE OneView management, and you can automatically configure SAN zoning through server profile volume attachments to mitigate configuration errors. Storage integration with server profiles saves your time and makes you more productive. Switched fabric, direct-attach (FlatSAN), and virtual SAN topologies are supported to provide dynamic connectivity between HPE OneView managed servers and HPE 3PAR StoreServ Storage systems. HPE OneView discovers the SAN paths and provides connectivity services for the following infrastructures. HPE 3PAR StoreServ Storage systems connected directly to an enclosure using FC HPE 3PAR StoreServ Storage systems connected to an HPE B-series FC SAN configuration –

SANs managed through the HPE B-series SAN Network Advisor software

HPE 3PAR StoreServ Storage systems connected to a Brocade FC SAN configuration –

||||||||||||||||||||

SANs managed through Brocade Network Advisor software

HPE FlexFabric 5900 AF/CP switches, Cisco MDS series switches, Cisco Nexus 5000 and 6000 series switches, and Brocade switches With HPE OneView, advanced automation enables an IT generalist to define and provision storage volumes, automatically zone the SAN as part of the provisioning process, and attach the volumes to server profiles. HPE OneView storage automation makes businesses more responsive, secure, and efficient. HPE 3PAR StoreServ Storage is fully integrated with HPE OneView server profiles for automated, policy-driven rollout of enterprise-class storage resources. After the storage has been rolled out, you can select an HPE 3PAR StoreServ volume in HPE OneView and create a snapshot from that volume. Snapshots in OneView allow copy and provisioning access to nonstorage professionals such as database administrators, software developers, and test engineers working with systems. Users can restore their own copies of test data safely and quickly without relying on a storage administrator. They can easily replace and restore copies of their volumes by copying,

||||||||||||||||||||

promoting, and attaching their volumes to server profiles. This enables users to update specific snapshots with more recent snapshots, resulting in faster turnaround times for developers who need refreshed snapshots. This also alleviates the workload for storage administrators. Using HPE 3PAR StoreServ Storage within HPE OneView is as simple as selecting a storage template and a server profile. HPE OneView automation carves out the storage volume, zones the FC SAN, and attaches the storage to the server profile. After they are rolled out, the SAN resources are immediately exposed in the topology map. This includes multihop FC and FCoE (Fibre Channel over Ethernet) architectures. In HPE OneView, proactive alerts are provided when the expected and actual connectivity and states differ or when SAN health issues are immediately visible in the topology map. HPE OneView provides SAN configuration reports, which include guidance for SAN efficiency and help in resolving potential SAN issues before there is a business impact. The HPE OneView appliance monitors the health status of storage systems and issues alerts when there is a change in status. The appliance also monitors the connectivity status of storage systems. If the appliance loses connectivity with a storage system, an alert is displayed until connectivity is restored. The appliance attempts to resolve connectivity issues and clear the alert. If it cannot, you must use the Storage Systems screen to refresh the storage system manually and synchronize it with the appliance. The appliance also monitors storage systems to ensure that they are synchronized with changes to hardware and configuration settings. However, changes to storage systems made outside the appliance (such as changing credentials) might cause the storage system to lose synchronization with the appliance, in which case you must manually refresh the storage system. HPE OneView Partner Integrations: HPE OneView media includes the virtual management appliance needed for optimally manage your HPE Converged Infrastructure. HPE OneView also includes integrations that provide additional support for partner management platforms that extend the functionality of HPE OneView within the context of the partner ecosystem. The following partner integrations are available as part of each HPE OneView license: –

HPE OneView for VMware vCenter and Log Insight

–

HPE OneView for Microsoft System Center

||||||||||||||||||||

HPE OneView provides a right-to-use (RTU) for the HPE OneView for VMware vCenter integration. This integration is a separate download and integrates HPE Technet24

||||||||||||||||||||

||||||||||||||||||||

OneView capabilities into VMware vCenter Server with viewing from the VMware console. HPE OneView provides an RTU for the HPE OneView for VMware vRealize Operations Manager and HPE OneView for VMWare vRealize Log Insight integration. These integrations are separate downloads. The integrations deliver deep analytics and troubleshooting. HPE OneView provides an RTU for the HPE OneView for Microsoft System Center integration. This integration is a separate download and integrates HPE OneView capabilities into Microsoft System Center.

HPE OneView for VMware vCenter With HPE OneView for VMware vCenter, HPE Converged Infrastructure management features are integrated directly into VMware vCenter. This allows administrators to use a familiar VMware management tool to provision, monitor, update, and scale HPE compute, storage, and network resources without having to leave the vCenter console. Rather than view physical and virtual infrastructures as two distinct entities, customers can use HPE OneView for VMware vCenter, as shown in 7-38, to manage both environments as one.

Figure 7-38 HPE OneView for VMware vCenter This integration simplifies everyday management tasks. Using wizards, administrators can deploy a VMware vSphere cluster in five easy steps and easily create storage volumes and vSphere datastores. HPE OneView makes the delivery and maintenance of IT services fast, cost effective, and reliable. The visual mapping of virtualized

||||||||||||||||||||

workloads to physical resources makes it possible to troubleshoot network problems in minutes instead of hours. HPE OneView for VMware vCenter is a single integrated application that customers can use to manage HPE ProLiant servers and storage systems. The application consists of two primary modules: HPE OneView for VMware vCenter Server Module provides: –

Comprehensive system health and alerting in VMware vCenter Server

–

Deep levels of HPE server, blade, firmware, and infrastructure inventory and management

–

End-to-end view of the HPE Virtual Connect networking environment

–

One-click host- and cluster-level networking configuration

–

Remote control and deployment of servers and operating systems

–

Ability to create or grow VMware clusters from HPE OneView profiles

HPE OneView for VMware vCenter StoreFront Module (formerly the Storage Module) allows users to: –

Monitor health, configuration, and capacity of HPE Storage in VMware vCenter Server.

–

Visualize the relationship between the VMware VMs, ESX servers, and HPE Storage arrays.

–

Create, expand, or delete datastores on HPE arrays.

–

Build a VM from a template on HPE arrays.

–

Include vSphere storage APIs for Storage Awareness for HPE arrays.

–

Support for HPE ProLiant and HPE Synergy Gen10 Servers.

–

Support for HPE Synergy Image Streamer for patching ESXi cluster images.

||||||||||||||||||||

HPE OneView for VMware vCenter administration HPE Converged Infrastructure management features are integrated directly into VMware vCenter. Allows administrators to use a familiar VMware management tool to provision, monitor, update, and scale HPE compute, storage, and network resources without having to leave the vCenter console

Technet24

||||||||||||||||||||

Simplifies everyday management tasks using visual mapping of virtualized workloads to physical resources HPE OneView for VMware vCenter ensures that VMware vCenter Server customers can view hardware-specific configuration and fault information from the console. From the Insight Software tab, VM administrators can view ProLiant server hardware and BladeSystem c3000 and c7000 enclosure configuration and health information at summary or detailed levels. Starting with 8.0 release, HPE OneView for VMware vCenter is available as an appliance only. It cannot be installed on Windows Platform. You can migrate the configuration information from the existing Windows Platform to the appliance with the help of Migration Tool. For more information about using the Migration Tool, see the HPE OneView for VMware vCenter Installation Guide. HPE OneView for VMware vCenter Server Module—Adds HPE ProLiant and HPE BladeSystem hardware monitoring into the HPE OneView for VMware vCenter console. Provides server hardware management capabilities, including comprehensive monitoring, firmware update, vSphere/ESXi image deployment, remote control, end-toend monitoring for Virtual Connect, and power optimization for HPE servers in the VMware environment. Important The HPE OneView for VMware vCenter Server Module and the HPE OneView for VMware vCenter Storage Module are no longer a separate installation. However, the HPE OneView for VMware vCenter Server Module services do not activate unless credentials are provided.

HPE OneView for VMware vCenter modules HPE OneView for VMware vCenter Server Module: –

System health and alerting in VMware vCenter Server

–

HPE server, blade, firmware, and infrastructure inventory and management

–

End-to-end view of Virtual Connect networking

HPE Storage Module for vCenter: –

||||||||||||||||||||

Enhances VMware functionality by providing details about the disks presented to the virtual environment.

||||||||||||||||||||

–

||||||||||||||||||||

Improves visibility of volumes.

The Storage Module for vCenter enhances VMware functionality by providing details about the disks presented to the virtual environment. This improved visibility helps administrators make better-informed decisions when designing, deploying, maintaining, and troubleshooting a virtual environment. For example, an administrator can use the information provided by the Storage Module for vCenter to do the following: Ensure that mission-critical VMs reside on storage provided by an HPE StoreVirtual storage systems meeting established replication requirements. Ensure that a production database application is located on the HPE StoreServRAID 5 storage volumes that are replicated to a remote datacenter. Preserve storage resources when placing a temporary test database on HPE MSA 2040/1040 storage volumes. From the vSphere Web Client, the Storage Module for vCenter enables you to do the following: Monitor the status and health of HPE storage systems. Understand the relationships between physical storage and virtual objects. VM—VM to datastore mapping and VM to raw device mapping. Datastore—Datastore to physical device mapping and datastore to VM mapping RDM —RDM to physical device mapping and RDM to VM mapping. Manage LUN (logical unit number) and volume connections from VMs, datastores, and hosts (ESX servers) to HPE storage systems. Provision storage on supported HPE storage systems. Take advantage of VASA storage API features (vSphere APIs for Storage Awareness). The Storage Module for vCenter supports the VASA storage API from VMware. The VASA API provides visibility into the physical storage infrastructure through vCenter. VASA capabilities are supported for HPE 3PAR StoreServ, HPE StoreVirtual, and HPE MSA 2040/1040 storage systems.

HPE OneView for VMware vCenter with Operations Manager and Log Insight HPE OneView for VMware vCenter with Operations Manager and Log Insight seamlessly integrates the manageability features of HPE Synergy, HPE ProLiant Server, Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE BladeSystem, HPE Virtual Connect, and HPE Storage with VMware solutions. You gain deep insight and control of virtualized HPE Converged Infrastructure environments while reducing the time it takes to make important changes, increase capacity, or manage planned and unplanned downtime. When used with the automation power of HPE OneView, best practices for a converged infrastructure can be defined once and reused many times to provision an entire cluster with compute and storage fully configured in five easy steps. Integrations with VMware vCenter® Operations Manager and Log Insight deliver powerful analytics and deeper troubleshooting tools to your VMware administrators. HPE OneView for VMware vRealize Log Insight (also known as HPE OneView for vRealize Log Insight) provides log aggregation and indexing with search and analytics capabilities. HPE OneView for vRealize Log Insight collects, imports, and analyzes logs to provide information related to systems, services, and applications.

HPE OneView for Microsoft System Center The Microsoft System Center family of integrated solutions enables data center managers to optimize resources, improve visibility of IT assets and issues, meet service levels, increase efficiencies, and decrease costs. HPE has delivered core management packs for the Microsoft System Center ecosystem for years, and with HPE OneView for Microsoft System Center, HPE and Microsoft customers can take advantage of in-depth infrastructure monitoring and management directly from the Microsoft System Center consoles.

HPE OneView for Microsoft System Center 2012 The health dashboard, as shown in Figure 7-39, is a Windows Presentation Foundation (WPF) user control dashboard. The health dashboard provides insights into the System Center Operations Manager (SCOM) data. You can retrieve information from the SCOM database through SCOM SDK (Software development kit). Note WPF is a graphical subsystem by Microsoft for rendering UIs in Windowsbased applications. (March 2018, Wikipedia.org)

||||||||||||||||||||

||||||||||||||||||||

Figure 7-39 Health dashboard information Additionally, the dashboard provides a graphical health representation of the following resources: HPE OneView Appliance, Server Hardware, Enclosure group, Server Profiles, Enclosures, Storage Pool, Storage Volume Storage System, and Drive Enclosures. To obtain metrics about an individual resource, click the resource to drill down further. Each resource Health GUI displays the following information: The total number of discovered instances The health status of the discovered instance in terms of percentage. The status icons (Error, Unknown, Warning, and healthy) under each resource groups the number of discovered instances along with their respective health status. HPE OneView for Microsoft System Center supports: SCOM System Center Virtual Machine Manager (SCVMM) HPE ProLiant and HPE Synergy Gen10 Servers Support for HPE OneView 3.1 Support for HPE Service Pack for ProLiant (SPP) 2017.07.0

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE OneView for Microsoft System Center provides seamless integration of the unique ProLiant and BladeSystem manageability features into Microsoft System Center. These extensions deliver comprehensive system health and alerting, driver and firmware updates, operating system deployment, detailed inventory, and proactive VM management in a flexible, easy-to-install package. HPE OneView for Microsoft System Center is backed by HPE service and support, ensuring timely service for customers. By integrating the server management features of ProLiant and BladeSystem into Microsoft System Center consoles, administrators can gain greater control of their technology environments. This provides consistency for software deployment and updates, which enables a faster response in the event of server failure and reduces the risk of downtime. HPE OneView for Microsoft System Center extends the best-managed ProLiant experience for customers who have standardized on a Microsoft System Center management platform. With HPE OneView for Microsoft System Center, customers can: Prevent problems from occurring with proactive monitoring and management. Quickly and reliably deploy bare-metal servers. Ensure consistency and maximize uptime with simplified driver and firmware updates. Reduce unplanned downtime with proactive VM movement. Avoid learning how to manage a separate tool because it is fully integrated into the system.

Support for System Center Virtual Machine Manager HPE OneView for Microsoft System Center provides end-to-end fabric visualization for virtualized environments using HPE Synergy or HPE Virtual Connect to view from VMs to the edge of the network configuration information. Provides enhanced provisioning using HPE OneView server profiles to deploy Hyper-V hosts consistently and reliably, including configuration of Windows® networking, HPE Virtual Connect, and shared SAN storage. Facilitates consistency and improved uptime with a simplified driver and firmware updates via a rotating, automated workflow for Hyper-V clusters using the HPE ProLiant Updates Catalog. Automates HPE storage management via SMI-S and provides an integrated view of VMs and associated storage resources, including end-to-end relationships between VMs

||||||||||||||||||||

||||||||||||||||||||

and storage with context sensitive, in-depth information for volumes and their business copies. Provides simplified integration with software-defined storage by transforming the existing server capacity into virtual storage through the integrated HPE StoreVirtual VSA deployment wizard for Hyper-V hosts and clusters.

Support for System Center Configuration Manager (SCCM) HPE OneView for Microsoft System Center provides quick and reliable Windows deployment to bare-metal HPE servers, including predeployment hardware and Basic Input/Output System (BIOS) configuration, and post-OS driver and agent installation. Verifies consistency and improves uptime with simplified Windows driver and firmware updates using the HPE ProLiant Updates Catalog. The HPE Fabric Management Add-in for System Center is part of HPE OneView SCVMM Integration Kit. It enables enhanced integration of SCVMM and HPE Storage providing a single pane-of-glass monitoring and provisioning of physical, logical, virtual, and cloud infrastructure. It automates HPE Storage management and provides an integrated view of VMs and associated storage resources. This provides enhanced monitoring, provisioning, and control of HPE Storage with following benefits: Simplistic relationships on dashboard info for VMs, Hyper-V hosts, volumes, and pools Simplified deployment of multiple VMs on Hyper-V hosts and clusters Unified monitoring of OneView storage resources Provides end-to-end relationship between VMs and HPE Storage with context-sensitive in-depth info of volumes and their business copies Flexibility in managing capacity requirements by expanding LUNs Enables monitoring complex System Center infrastructure by integrating with SCOM management pack and listing storage alerts Manages HPE StoreVirtual VSA appliances Licensing

Storage management with HPE OneView for Microsoft System Center can be used at no charge. Licensing is enabled as part of the download and installation sequence. Server management with HPE OneView for Microsoft System Center is licensed with OneView and HPE Insight Control. New advanced provisioning capabilities are

Technet24

||||||||||||||||||||

||||||||||||||||||||

available for OneView licensed host only. HPE OneView for Microsoft System Center 8.3: HPE OneView Storage System Management Pack (part of HPE OneView SCOM Integration Kit) Supports OneView 3.0 and 3.1. IPV6 support for SCOM. HPE Storage Management Pack for System Center (part of HPE Storage SCOM Integration Kit) Supports latest StoreVirtual OS 12.7 and 13.5. Supports new 3PAR StoreServ 9450. Supports new 3PAR StoreServ 20800 R2/20840 R2/20850 R2. Supports new MSA 2050. Monitoring of File Persona components for HPE 3PAR StoreServ. HPE Fabric Management Add-in for System Center (part of HPE OneView SCVMM Integration Kit) Supports OneView 3.0 and 3.1. VM creation wizard is modified to replace OneView resources with HPE 3PAR StoreServ and HPE StoreVirtual storage for provisioning. Supports new HPE 3PAR StoreServ 9450. Supports new HPE 3PAR StoreServ 20800 R2/20840 R2/20850 R2. Supports new HPE StoreVirtual OS 12.7 and 13.5. Note For more detailed information on the partner integrations, visit: HPE OneView for VMware vCenter, Operations Manager and Log Insight www.hpe.com/products/ovvcenter HPE OneView for Microsoft System Center www.hpe.com/products/ovsc

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check HPE OneView supports KVM. True False

||||||||||||||||||||

||||||||||||||||||||

Activity: What is new with OneView 4.0? Instructions: Watch the “HPE OneView—What’s new with 4.0” at: https://h20195.www2.hpe.com/V2/GetDocument.aspx?docname=A00029858ENN. Notes:

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check Write a summary of the key concepts presented in this chapter.

||||||||||||||||||||

||||||||||||||||||||

Summary HPE 3PAR manageability and monitoring is provided through many different options. Managing and monitoring StoreOnce from the management software can be completed through the GUI or the CLI. Nimble Setup Manager is a Windows Utility that uses automatic array discovery. HPE offers a wide range of monitoring tools to help customers improve IT processes, such as Analytics for HPE OneView, HPE OneView for VMware vCenter, and HPE OneView for Microsoft System Center.

Technet24

||||||||||||||||||||

||||||||||||||||||||

8

Upgrading, Optimizing, and Tuning HPE Storage Solutions

||||||||||||||||||||

LEARNING OBJECTIVES After completing this chapter, you should be able to: Upgrade the following systems: –

Hewlett Packard Enterprise (HPE) 3PAR StoreServ system

–

HPE Nimble system

–

HPE StoreOnce System

||||||||||||||||||||

Optimize HPE Storage systems Tune HPE 3PAR StoreServ Update HPE 3PAR operating system Update HPE 3PAR Service processor

Technet24

||||||||||||||||||||

||||||||||||||||||||

Prelearning check You are looking to add some new drives in your 3PAR StoreServ 8400 array. To maintain best practices, what would be the minimum upgrade drive quantity?

||||||||||||||||||||

||||||||||||||||||||

Upgrading the HPE 3PAR StoreServ system HPE provides hardware replacements and upgrades as well as HPE 3PAR OS software updates and patches for HPE 3PAR StoreServ customers. Component firmware at all levels can be upgraded dynamically (all together or individually), including the disk firmware, because all disk devices in the HPE 3PAR StoreServ system are dual ported. Intuitive functionality such as checkhealth functionality helps confirm that an upgrade can be completed successfully and without risk when performed before an HPE 3PAR OS upgrade. These functions check firmware levels, system services, and system status in terms of any degraded or failed components that might need replacing. In addition, the health check can be executed on an ongoing basis to obtain comprehensive information on the health of the system. Proactive system monitoring and support services can also be used to remotely apply software upgrades and dynamically examine a system at the customer’s request, helping ensure successful upgrade experiences when change control windows are limited.

High availability Best practice: HPE encourages all HPE 3PAR StoreServ Storage customers’ upgrade to the latest recommended HPE 3PAR OS. Upgrading to the most current HPE 3PAR GA OS ensures that the storage system benefits from the ongoing design improvements and enhancements. For customers participating in the Get 6-Nines Guarantee Program, the program will identify the latest HPE 3PAR OS version that is covered under the Guarantee program. Best practice: Size the system appropriately so that all workloads and applications dependent on the HPE 3PAR system can perform as needed under the conditions of a node being down. This might occur during an unplanned controller node failure or planned maintenance of a controller node. In no situation should the maximum limits of the system as defined in this document and product specifications be exceeded. In systems with 4 or more nodes, a resilience feature called Persistent Cache is automatically enabled. The Persistent Cache feature ensures that no storage controller node is placed into performance limiting “cache write thru” mode as a result of losing its partner in the node pair. Any node that loses its adjacent node can dynamically form a mirrored cache relationship with another storage controller node. This limits the performance impact of unplanned downtime or controller node maintenance.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Adding new components to an existing system Follow the general guidelines by completing the following: Check configuration guideline. Back up data. Check health of system. Prepare an electrostatic discharge (ESD) environment. Add new components. Configure new components. Check health of updated system. Tune or rebalance system. When performing any upgrade while the customer is concurrently using the system, exercise extreme caution because any incorrect action taken during an upgrade might cause the system to fail. Before performing any upgrade or expansion, verify with the system administrator if a complete backup of all data on the storage system has been performed. The general recommended order of installing components is shown in Figure 8-1.

||||||||||||||||||||

||||||||||||||||||||

Figure 8-1 Adding new components to an existing system Make sure that the system is fully operational and does not have any failed components.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Use the checkhealth command to verify the operational state. The checkhealth command checks the status of system hardware and software components and reports any issues. checkhealth [|] [options]: –list Lists all components that checkhealth can be run on. –quiet Does not display which component is currently being checked. –detail Displays detailed information regarding the status of the system. indicates the component to check. Use the -list option to get the list of components (alert, cage, date, ld, license, network, node, pd, pdch, port, rc, snmp, task, vlun, and vv). Notice Storage systems are susceptible to failure caused by ESD. Electrostatic charges can accumulate on people and furniture because of direct contact with floor coverings or movement while in contact with furniture coverings. Discharge of static electricity to a metal surface on server cabinets can interfere with system operation and cause discomfort to anyone who comes in contact with it. To prevent damage from ESD, you must wear an ESD wrist-grounding strap during installation and maintenance of the storage system. Place all components during removal and replacement on an ESD-safe work surface.

Controller node upgrade Installing additional controller nodes, as shown in Figure 8-2, enhances performance and increases maximum storage capacity. Considerations include: You cannot upgrade a x200 storage system to a x4x0. You cannot upgrade a x400 storage system to a x450. Only a two-node x4x0 storage system can be upgraded to a four-node system. The x200 and x4x0 system nodes cannot be mixed. Older controllers must be changed for the same model. Only new nodes can be used to replace existing nodes or for upgrades in a x4x0. Nodes

||||||||||||||||||||

||||||||||||||||||||

cannot be moved from one system and installed in another. Nodes in a node pair must have identical HBA/CNA/NIC configurations. All upgrades must be performed by qualified personnel.

Figure 8-2 Controller node upgrade

Upgrading online controller nodes with Persistent Ports When you are upgrading controller node software, the HPE 3PAR StoreServ system enables you to upgrade one controller node at a time, pausing between nodes to confirm that multipathing is active. HPE 3PAR StoreServ Persistent Ports is a feature of HPE 3PAR OS that provides resiliency and decreases the need for host multipathing

Technet24

||||||||||||||||||||

||||||||||||||||||||

software. With Persistent Ports in SAN connect environments, the Fibre Channel SAN switch that is connected to the HPE 3PAR StoreServ Storage system supports N_Port ID virtualization (NPIV) capability. This means you can perform software upgrades on a node-by-node basis without any reliance on host multipathing software, thereby avoiding any path failovers, as shown in Figure 8-3.

Figure 8-3 Persistent Ports The Persistent Ports feature also works with integrated Small Computer System Interface (iSCSI) and FCoE (Fibre Channel over Ethernet) ports and allows nondisruptive and online upgrades to take place completely transparent to hosts. In addition to enabling nondisruptive and online software upgrades, Persistent Ports also protects applications against any possible outage because of the loss of a controller node. Regardless of the cause of this loss—planned upgrade or maintenance, power failure, and other reasons—the Persistent Ports feature allows seamless failover of all ports on the impacted node to designated partner ports on a different node.

||||||||||||||||||||

||||||||||||||||||||

This is seamless to hosts, which continue to use the original ports assigned and I/O is transparently rerouted through the partner port. This helps organizations maintain uninterrupted service levels for their applications running on HPE 3PAR StoreServ Storage and protects against any service-level penalties that might apply. For Fibre Channel–attached hosts, Persistent Ports functionality also provides transparent and uninterrupted failover in response to “loss_sync” events, which might be triggered as the result of a failure between an array HBA and the switch. Persistent Ports configuration considerations Persistent Ports requires that corresponding “Native” and “Guest” host ports on a node pair be connected to the same Fibre Channel (FC) fabric or IP network, and the switches they are connected to must support and be configured for NPIV in the case of FC and FCoE. This means that for a minimum configuration to provide Persistent Ports functionality, where the node pair is connected to redundant FC SAN fabrics, each node in a node pair must have at least two FC host ports cabled with one port connected to each fabric.

Drive enclosure upgrade The default provisioning for all HPE 3PAR StoreServ Storage systems includes high availability across disk chassis (or enclosures). (This is the high-availability [HA] Cage or HA enclosure mode.) This default helps ensure data availability in case of the loss of an entire enclosure. Even in the event of an entire disk enclosure loss, the hosts do not lose access to the provisioned logical unit numbers (LUNs) and applications still stay online. Because it is likely that not all data is of the same importance, this functionality is dynamically configurable per LUN. The minimum disk availability is magazine level. As the system scales, the availability level per LUN can be nondisruptively changed to disk-shelf availability as desired. Considerations include: Two types of drive enclosures are used for expansion. The number of drive enclosures is determined by the desired Redundant Array of Inexpensive Disk (RAID) set size and HA cage protection requirements. When adding both 2U and 4U drive enclosures, they should be mixed on SAS (Serial Attached SCSI) chains (DP-1 and DP-2), added in pairs across controller node pairs on a four-node system, and balanced across SAS ports on each controller node pair. Distribute drive enclosures evenly between DP-1 and DP-2, as shown in Figure 8-4.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 8-4 Drive enclosure upgrade

Disk chassis availability HA cage Default provisioning for all HPE 3PAR StoreServ systems includes high availability across disk chassis. This ensures data availability in case an entire enclosure is lost. No data ever pinned in cache

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ Storage supports resilient batteries to help ensure that the contents of write cache are written to specific and secure disks if a site power failure occurs. There is no pinning of data in cache, which puts data at risk when power is not restored in a specified period of time. Because cache data is written to disk, there is no limit on the time to data recovery for cases where power is down for an extended period. After power is restored, the cached data that was written to the secure disks is restored to its original location as part of the power-fail recovery process. Rapid recovery from failed disk devices In general, the most common hardware failure for storage arrays occurs within the disk devices. In the event of a disk failure, it is imperative that rebuild takes place quickly and with little or no impact to system performance. HPE 3PAR StoreServ Storage systems use unique capacity virtualization technology to help ensure that all potentially impacted LUNs associated with a physical disk device failure are rebuilt from a degraded mode to full parity protection in the fastest possible time. On the HPE 3PAR StoreServ Storage system, all disk devices are virtualized into chunklets. The system tracks utilization and writes allocation for each chunklet at all times. All disk devices have a number of spare chunklets that are reserved for sparing purposes. Note Physical disks are divided into chunklets. Each chunklet occupies contiguous space on a disk. Chunklets on HPE 3PAR StoreServ 10000 and 7000 and later storage systems are 1 GB. The system uses a global spare chunklet pool that spans all physical drives in the system instead of more limited and traditional hot-sparing concepts. This global pool comprises a specific amount of spare capacity, regardless of the size and type of disk devices in the system. If the system is upgraded, an embedded algorithm helps ensure that the global chunklet sparing pool grows in proportion to the capacity installed. If a disk device fails, an automated RAID rebuild occurs to spare chunklets distributed across all disk devices, resulting in an expedited and extremely rapid rebuild window. Traditional storage rebuild models involve a few disks in a disk RAID group rebuilding to a hot spare, which can take significantly longer for RAID rebuild and result in I/O hotspots. The massive parallelism resulting from global sparing helps ensure limited impact on application performance. In addition, HPE 3PAR StoreServ Storage disk rebuilds are thin-provisioning aware, so only chunklets that have been written to are rebuilt. This

Technet24

||||||||||||||||||||

||||||||||||||||||||

also accelerates the rebuild time to full parity. For example, if a 2 TB disk fails but only 300 GB of written data resides on it, only 300 GB needs to be rebuilt to return any affected LUNs back to full RAID parity. The system checks the World Wide Name (WWN) of each replaced disk device to help ensure that the failed disk has not accidentally been reinserted, thus minimizing extended service times.

Upgrading storage capacity If future capacity upgrades are expected, include enough drive enclosures in the initial configuration so that there are some empty bays in each enclosure after all the drives have been added. The number of drive enclosures attached to a specific node pair should be determined by the desired RAID set size and HA Cage protection requirements. Drive enclosures should be added and configured to meet the HA Cage protection requirement for a specific node pair, and the customer’s RAID set requirement should also be considered. When adding or upgrading drive enclosures, keep these best practices in mind: Minimum upgrade drive quantity is four drives per node pair or two drives per enclosure (counting enclosures that support the drive’s form factor), whichever is larger. Add upgrade drives in identical pairs to an enclosure. When adding upgrade drives, add equal numbers of drives to all enclosures that support the form factor of the drives. When upgrading capacity in a four-node configuration, add equal quantities and types of drives to both node pairs. Run Autonomic Rebalance (also known as tunesys) after adding drives.

Recommended configuration rules If future capacity upgrades are expected, include enough drive enclosures in the initial configuration so that there are some empty bays in each enclosure after all the drives have been added. For all HPE 3PAR StoreServ models, only 2.5-inch drives can be installed in controller node enclosures and HPE 3PAR StoreServ 8000 2U drive enclosures. The 3.5-inch drives can be installed only in HPE 3PAR StoreServ 8000 4U drive enclosures. Each drive enclosure holds up to 24 drives in 2U for small form factor (SFF) drives and 4U for large form factor (LFF) drives. Upgrade options for enclosures and drives for

||||||||||||||||||||

||||||||||||||||||||

StoreServ 8000 are shown in Figure 8-5. Some HPE 3PAR StoreServ models require drive enclosures.

Figure 8-5 Upgrade options for enclosures and drives for StoreServ 8000

Drive and enclosure upgrades For all drive enclosures, the slots should be balanced. Best practice is to add multiples of two identical drives to each enclosure on the array at the same time for optimal performance and availability. Controller node enclosures include 24 drives attached to the SAS port–labeled DP-1. When you are adding both 2U and 4U drive enclosures, you should mix them on SAS chains (DP-1 and DP-2), added in pairs across node pairs on a four-node system, and balanced across SAS ports on each controller pair. Drive enclosures between DP-1 and DP-2 should be distributed to achieve maximum balance across the ports. For example, if two drives are added to DP-1, two drives should be added to a drive enclosure attached to DP-2. Drives should be added so that all enclosures are balanced with an even number of drives in each enclosure. In SFF enclosures, drives must be added from left to right, with no gaps. Unused slots must contain drive blanks. In LFF enclosures, drives must be added with no gaps, starting from the bottom of the left column and filling this column from bottom to top. Populate the remaining columns from bottom to top and from left to right. Columns might contain a mix of 7.2K hard disk drives (HDDs) and solid-state drives (SSDs); unused slots must contain drive blanks.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Note The specifications for the 8000, 9000, and 20000 models are 16 Gb Fibre Channel and 12 Gb SAS only. Minimum drive configurations for HPE 3PAR StoreServ 8000 models are: Minimum of six SSDs per node pair Minimum of four SSDs per node pair if used for Adaptive Flash Cache Minimum of eight 10K/15K HDDs per node pair Minimum of 12 7.2K HDDs per node pair For all drive enclosures, the proper system temperature must be maintained. To ensure proper thermal control, blank filler panels must be installed in any slots without drives. Allocating and loading order for 2.5-inch SFF disk drives—For a node or an HPE 3PAR StoreServ 8000 2U drive enclosure, drives must be added in identical pairs, starting from slot 0 on the left and filling to the right, leaving no empty slots between drives. The best practice when upgrading or building a system is to add the same number of identical drives to every drive enclosure in the system, with a minimum of two disk drive pairs per drive enclosure. Allocating and loading order for 3.5-inch LFF disk drives—Columns of drives in an HPE 3PAR StoreServ 8000 4U drive enclosure must be of the same device type (NL or SSD). NL disk drives and SSDs must not be mixed in the same column. For a 4U HPE 3PAR StoreServ 8000 drive enclosure, drives must be installed in identical pairs, starting from the slot at the bottom of the left column and filling up with drives of the same type, leaving no empty slots between drives in the column. Populate drives in the columns from bottom to top and from left to right. The best practice while upgrading or building a system is to add the same number of identical drives to every HPE 3PAR StoreServ 8000 drive enclosure in the system, with a minimum of two drives added to each LFF chassis. The minimum supported upgrade for an HPE 3PAR 8000 with one or more expansion drive enclosures is two identical drives added to adjacent slots in the same column of the same chassis. If there is more than one expansion enclosure in the system, add the drives in pairs so they are balanced across the device ports, and then add them to enclosures on the same port. The LFF drives added to the HPE 3PAR StoreServ 8000 drive enclosure on each node pair should be balanced across node pairs, then across device ports on each node pair, and finally by enclosures on the same port.

||||||||||||||||||||

||||||||||||||||||||

Drive enclosure expansion limits The disk drives in the node enclosure are connected internally through DP-1. Expansion limits are as follows: An HPE 3PAR StoreServ 8200 array can support up to nine drive enclosures. An HPE 3PAR StoreServ 8400 array can support up to 22 drive enclosures. An HPE 3PAR StoreServ 8440 array can support up to 38 drive enclosures. An HPE 3PAR StoreServ 8450 array can support up to 18 drive enclosures. An HPE 3PAR StoreServ 9450 array can support up to 48 drive enclosures. An HPE 3PAR StoreServ 20450 array can support up to 40 drive enclosures. An HPE 3PAR StoreServ 20850 array can support up to 80 drive enclosures. The recommended minimum number of drive enclosures in the 8000 series arrays is four drive enclosures per pair of controller nodes. Note Detailed information about number of drive enclosures is available in the QuickSpecs document for each HPE 3PAR StoreServ model. To access QuickSpecs, open the hyperlink. http://h41370.www4.hpe.com/quickspecs/overview.html

Information about disk drive upgrades You can install additional disk drives to upgrade partially populated drive enclosures. Important If the HPE 3PAR StoreServ is enabled with the Data at Rest (DAR) encryption feature, only use the self-encrypting drives (SEDs). Using a non-SED might cause errors during the upgrade process. SSDs have a limited number of writes that can occur before reaching their write endurance limit. This limit is generally high enough, so wear-out does not occur during the expected life span of an HPE 3PAR StoreServ system for the majority of configurations, I/O patterns, and workloads. HPE 3PAR StoreServ tracks all writes to SSDs and can report the percent of the total write endurance limit that has been used. Technet24

||||||||||||||||||||

||||||||||||||||||||

This allows any SSDs approaching the write endurance limit to be proactively replaced before they are automatically spared out. An SSD reaches the maximum usage limit when it exceeds its write endurance limit. Following the product warranty period, SSDs that have exceeded the maximum usage limit will not be repaired or replaced under HPE support contracts. The first expansion drive enclosure added to a system must be populated with the same number of disk drives as the node enclosure. Disks must be identical pairs. The same number of disk drives and type should be added to all the drive enclosures in the system. The minimum addition to a two-node system without expansion drive enclosures is two identical disk drives. The minimum addition to a four-node system without expansion drive enclosures is four identical disk drives.

Allocating and loading guidelines—SFF drives Loading guidelines for SFF drives, shown in Figure 8-6, include: Load SFF drive types (FC, NL, and SSD) in pairs. Load starting from slot 0 on the left and filling the slots sequentially to the right. Load all of the same drive types in adjacent slots before loading the next drive type. Try to allocate the same number of drive-type pairs to every enclosure in the system. Each add-on enclosure requires a minimum of 4 drives.

Figure 8-6 An SFF enclosure loaded with two pairs of FC drives, three pairs of NL drives, and two pairs of SSDs

Allocating and loading guidelines—LFF drives Loading guidelines for LFF drives, shown in Figure 8-7, include: Load LFF drive types in pairs.

||||||||||||||||||||

Load starting from slot 0 on the bottom of the left column and filling the slots moving upward in the column, before moving to the bottom of the next column to the right. –

||||||||||||||||||||

Columns might contain a mix of NL HDDs and SSDs.

Load all the same drive types in adjacent slots before loading the next drive type. Try to allocate the same number of drive-type pairs to every enclosure in the system.

Figure 8-7 An LFF enclosure loaded with one pair of SSDs and three pairs of NL drives If there is more than one expansion enclosure in the system, load the drives in pairs, so they are balanced across the enclosure node connections DP1 and DP2, and then added to enclosures in a vertical column.

Allocating and loading guidelines—Avoid practices Avoid loading the SFF drive in the manner shown in Figure 8-8, because this example demonstrates an unbalanced load, due to the NL drives not being installed in even pairs. Avoid having an odd number of NL drives in both cages. Avoid leaving any empty slots between drives. Avoid an unbalanced number of same-type drives. Avoid loading same-type drives in slots that are not adjacent to each other. Avoid filling an entire drive enclosure when additional enclosures are available. Leaving empty slots in each enclosure allows for future upgrade flexibility to add a new drive type and license features.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 8-8 Allocating and loading guidelines—Avoid practices

Disk drive upgrade There are five steps for adding hard drives, as shown in Figure 8-9: Check the initial status—Under Systems, select Physical Disks, and in the right pane, select the Physical Disks tab. Insert hard drives—For example, add two disk drives to each of the three enclosures. Check the status—The display refreshes periodically, and you should see the inserted disk drives as New in the State column. They are ready to be admitted into the system, which occurs automatically.

||||||||||||||||||||

||||||||||||||||||||

Figure 8-9 Disk drive upgrade Within six minutes (depending on system load and upgrade size), the state of the new disk drives changes to Normal, and the system prepares the chunklets for use. Output indicates that the six disk drives still have normal and spare chunklets to be initialized. Chunklet initialization can take several hours to complete, and the output of the available capacity is displayed. Note The system can be used normally, but newly added capacity must be initialized before it can be allocated.

Check progress—On the Physical Disks tab, in the drop-down list, select Chunklet Usage.

Technet24

||||||||||||||||||||

Complete the upgrade—To complete the upgrade, you must add the expansion drive enclosures by following these steps: a. Install the expansion drive enclosure. b. Install the disk drives, and cable the enclosures to each other by using SAS cables. c. Install the 580 W power and cooling modules (PCMs) into the enclosure. d. Install the power cables to the PCMs and press ON (the power switch). e. After you have completed the physical installation of the enclosures and disk drives, cable the drive enclosure to the controller nodes. f. Verify that the upgrade is successful.

||||||||||||||||||||

Safe to remove The safe to remove feature currently supports 20000 StoreServ arrays only. A Safe to remove property has been added to the list pane Summary view for Controller Nodes, Drive Enclosures, and Physical Drives screens, as shown in Figure 8-10.

||||||||||||||||||||

||||||||||||||||||||

Figure 8-10 Safe to remove

Customer self-repair In the Systems menu Overview screen, you can confirm the health status of your system including Drives and Enclosures, as shown in Figure 8-11. Selecting the Degraded hyperlink will drill down into more details.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 8-11 HPE 3PAR StoreServ Management Console (SSMC) health monitoring

||||||||||||||||||||

||||||||||||||||||||

Learning check You are looking to add some new drives in your 3PAR StoreServ 8400 array. To maintain best practices, what would be the minimum upgrade drive quantity?

When you are about to perform an online controller node software upgrade, what hostbased software needs to be active?

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: HPE 3PAR StoreServ 9000 Storage Customer Self-Install video Instructions: Watch the first five minutes of the “HPE 3PAR StoreServ 9000 Storage Customer Self-Install video” at: http://www.hpe.com/support/3PAR9000CSIVideo. Note This video is an hour long, and you will just have a look at the first five minutes. You can come back to this at any time. Notes:

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE InfoSight and upgrading HPE Nimble systems

Figure 8-12 Upgrading HPE Nimble systems HPE InfoSight also takes the guess work out of planning. It accurately predicts future capacity, performance and bandwidth needs and right sizes your new infrastructure through app-centric modeling, as shown in Figure 8-12. Our forecasting models are statistically highly confident using autoregressive and Monte Carlo simulations taking into account irregular patterns that would be considered by an untrained eye as unpredictable. And planning for new infrastructure is a cinch, because InfoSight lets you model different scenarios, applications, users, and performance requirements. As new infrastructure is deployed, the application performance is monitored and the models keep getting refined. Note Monte Carlo simulation is a computerized mathematical technique that allows people to account for risk in quantitative analysis and decision making. The technique is used by professionals in such widely disparate fields as finance, project management, energy, manufacturing, engineering, research and development, insurance, oil and gas, transportation, and the environment. (March 2018, www.palisade.com)

||||||||||||||||||||

||||||||||||||||||||

Upgrading the HPE Nimble system When upgrade recommendations are present, they provide all the data needed to explain why the upgrade is needed, and specifically what is needed to address the issue. For all-flash arrays, upgrade recommendations can be related to controller CPU/memory upgrades and capacity upgrades. For adaptive flash arrays, additional recommendations include cache upgrades. All the data represented in the easy to read entries can be exported for more detailed analysis if needed. All upgrade recommendations consider the last four weeks of activity. In the case that multiple components of an array need to be upgraded, an arrow icon and the text upgrade recommended will appear below the specific graph (CPU, cache, or capacity).

What is in a AF & HF Array? Figure 8-13 shows a high-level overview of what is in a Nimble Storage HF-Series array and what can be connected to a controller shelf. There are three primary components in each Controller Head Shelf: HDDs and SSDs (located in the front of the chassis) Dual power supplies (located in the back of the chassis) Dual controllers (located in the back of the chassis)

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 8-13 AF & HF Arrays The Controller Head Shelf can then have the All-Flash Shelf and the Disk Expansion Shelves connected to it. Both the All-Flash Shelf and the Disk Expansion Shelves are optional.

Drive layout—HF20H, HF20, HF20C, HF40, HF60 The HF-Series chassis are 4U in height and support twenty-four 3.5” slots, which can be populated with 21 HDDs and three Dual Flash Carriers (DFCs) for a total potential of six SSDs. The minimum number of SSDs that can be configured with these arrays is three, which populates Bank A in the DFC. The chassis slots are numbered from left to right and top to bottom, as shown in Figure 8-14. Slots 1 through 3 contain the DFCs, and slots 4 through 24 are populated with disk carriers.

||||||||||||||||||||

Figure 8-14 HF-Series slots numbering Specifications include: 4U 24 chassis. 24 x 3.5” Slots carry 21x HDDs + 3x DFCs. HDDs: 18 + 3 RAID. –

New Nimble-branded HDD carriers.

DFCs: minimum of three SSDs in Bank A. –

||||||||||||||||||||

Bank B is available for cache upgrades.

Drive layout—DFC details The DFC is the same carrier found in the all-flash array. Each carrier has a Bank “A” and a Bank “B” supporting up to two SSDs as shown in Figure 8-15. Each bank has its own release latch, allowing users to remove a single SSD, while the DFC that contains both banks has a single latch that removes both bank slots (A and B).

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 8-15 DFC details The LED panels display the same LEDs that are found on all Nimble Storage Arrays.

Drive layout—HF201000H: A half-populated HF20 The first 11 HDDs are used to create a single RAID group. When the HF-20 is expanded, the second set of 11 drives creates a second RAID group that is added to the storage pool. The CS1000H is a CS1000 that is half populated. This configuration introduces a few differences from the other CS-Series chassis. The first difference is in the number of HDDs and SSDs supported. When half populated, it supports 11 HDDs and two DFCs with only slot A of the DFCs populated. These are located in the top half of the chassis. At a later time, customers can upgrade a HF20FP to a HF20FP (HF20 fully populated), as shown in Figure 8-16. Once upgraded, the CS1000FP supports 22 HDDs and two fully populated DFCs.

||||||||||||||||||||

||||||||||||||||||||

Figure 8-16 HF201000H drive layout The second difference is that, when fully populated, the HF20FP consists of two RAID stripes—one in the top half of the chassis consisting of 11 HDDs and the other in the bottom half of the chassis. This means that the capacities of a HF20H and HF20FP will be different than a HF20. Note The WebUI will display only HF20. To identify a HF20H or HF20FP, look at the system capacity or navigate to Manage → Array → [select array], and view the visual representation.

HF-Series connectivity options The HF-Series has a wide array of connectivity options available, as shown in Figure 8-17. Options range from the two default onboard ports up to six additional ports. The HF-Series can also be ordered with iSCSI or FC protocols.

Technet24

||||||||||||||||||||

Figure 8-17 HF-Series connectivity view CS-Series ports: By default, there are two onboard 10 Gbase-T ports. –

An array can be ordered with only these ports if desired.

Ethernet –

1 Gb and 10 Gb ports available for order

–

Optical available

–

Support for up to 12 ports

–

Quad port cards

Fibre Channel –

16 Gb FC

–

Support for up to 12 ports

–

Quad port cards

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Nimble HFx0 to fit Just like the AF-Series arrays, the HF-Series arrays can scale in many ways, as shown in Figure 8-18. Upgrade the amount of available cache or upgrade your controllers. Add expansion shelves or an All-Flash Shelf. Cluster four arrays together and expand/upgrade them as needed.

Figure 8-18 HF-Series scaling options

HFx0 controller upgrades In addition to upgrading cache, users can also upgrade HF-Series arrays to a higher model, as shown in Figure 8-19. This is done by replacing the controllers using the same procedure as with the AF-Series arrays. This upgrade is nondisruptive within

Technet24

||||||||||||||||||||

||||||||||||||||||||

HFx0 controller family. Be aware that cache upgrades might be required to meet the proper Memory to Flash Ratio (MFR).

Figure 8-19 HF-Series controller upgrade path Features: Nondisruptive controller upgrades within CSx000 controller family. Cache upgrades might be required to meet MFR. Timeless Storage supported at launch.

HF10 “half-pop” controller upgrades No further controller upgrades beyond HF40 are supported. Upgrading from the HF20C requires a minimum of 1440GB cache if the HDD size is 21TB. Capacity and controller upgrades are non-disruptive. Timeless Storage supported at launch.

Figure 8-20 HF controller upgrade path

||||||||||||||||||||

||||||||||||||||||||

ES3 Expansion Shelf

Figure 8-21 HF-Series expansion shelf Capacity of the HF-Series arrays can be scaled with the addition of one to six Gen2 expansion shelves.

Drive layout—ES3

Figure 8-22 ES3 drive layout From the front, the ES3 chassis looks identical to a HF-Series controller shelf and includes 21 HDDs and up to six SSDs, as shown in Figure 8-21and Figure 8-21. Technet24

||||||||||||||||||||

||||||||||||||||||||

Specifications include: 4U 24 chassis. 24 x 3.5” slots carry 21x HDDs + 3x DFCs. HDDs: 18 + 3 RAID. New Nimble-branded HDD carriers. DFCs: Bank A preconfigured with three SSDs, as shown in Figure 8-22. Bank B available for cache upgrades.

||||||||||||||||||||

||||||||||||||||||||

Upgrading HPE StoreOnce Systems StoreOnce Systems vary by design, so for each system, it is necessary to consult the correct Maintenance and Service guide. Components that can be upgraded on StoreOnce include: Software Firmware Capacity Nodes—on the high-end systems Capacity upgrade comes in the form of upgrade kits, for example: The HPE StoreOnce 5100 System (48 TB) Capacity Upgrade Kit (BB916A) is a 2U disk enclosure containing twelve 4 TB disks. It includes a hot spare disk. The StoreOnce configuration has a single I/O module (in the top location) plus a “blank” module. Components of the upgrade are shown in Figure 8-23.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 8-23 HPE 5100 Capacity Upgrade system components

HPE 5100 Capacity Upgrade system components

||||||||||||||||||||

||||||||||||||||||||

Note A 1-meter SAS cable is supplied with the product, as well as 0.5-meter and 2meter cables might be purchased. Cables over 2 meters are not supported. Many of the maintenance procedures require the system to be powered off. Advise users of the Maintenance window and when the HPE StoreOnce System will be unavailable. For more assistance, visit: https://sml-csr.ext.hpe.com

Firmware upgrades Many of the maintenance procedures require the system to be powered off. Advise users of the Maintenance window when the HPE StoreOnce System will be unavailable and use the StoreOnce Graphical User Interface (GUI) or Command Line Interface (CLI) to power off cleanly and then disconnect the system from the power supply. To power off: StoreOnce GUI: Select Maintenance from the Navigator and click Shutdown, as shown in Figure 8-24. StoreOnce CLI: Use the StoreOnce CLI command system shutdown.

Figure 8-24 HPE StoreOnce Firmware upgrades

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE StoreOnce 5100 System only: Any storage enclosure attached to the head server is automatically powered off at the same time. Storage enclosures do not have separate power on/off buttons but should be unplugged from the power supply before carrying out maintenance activities. To power on: HPE StoreOnce 5100 System only: Reconnect all storage enclosures to the power supply first. They are automatically powered on. Allow the power on to complete before powering on the head server; wait for the LEDs to show solid green. Press the Power On button on the HPE StoreOnce System to power up the appliance.

Upgrading firmware Performing firmware maintenance from the StoreOnce GUI: From the “Device Configuration” menu item in the Navigator, select Firmware. Click Scan to perform a check on the currently installed firmware versions for all components associated to the selected node. This will populate the firmware tabs and might take several minutes to run. Look for components that have a “Recommended Action” of either “Upgrade” or “Downgrade.” After you have run Scan or Scan All, components will automatically be selected for upgrade or downgrade. Note When the firmware updates are initiated, the backup devices will be taken offline. Before performing this step, check that there are no outstanding backup jobs pending. If there are only one or two, click in the appropriate check box to select them and click Update Selected, if there are many, click Update All. When all updates are completed, a reboot might be required.

||||||||||||||||||||

||||||||||||||||||||

Note It is a good idea to enable Remote Event Suppression mode when upgrading firmware and software. When enabled, Remote Event Suppression flags all outbound event messages to let the HPE Support team know that the system is undergoing some form of maintenance, such as network reconfiguration, that does not require a case to be raised with HPE Support. On the StoreOnce GUI, select Events in the Navigator and go to the Settings tab. Or use the StoreOnce CLI command system enable remoteeventsuppression "reason", as described in the HPE StoreOnce System CLI Reference Guide. The currently supported firmware component updates are embedded in the StoreOnce software. Therefore, to ensure that you are using the currently supported firmware component updates, always use the StoreOnce GUI or the StoreOnce CLI to check and update the Basic Input/Output System (BIOS) and hardware firmware components. Do not upgrade BIOS or hardware firmware components individually using downloads from the HPE Support website.

Multinode systems To update firmware using the StoreOnce CLI: Run the StoreOnce CLI command to show the current status of firmware # hardware show firmware <nodeX|storageX|switch> Note X is the number of the node or storage to which the command should be applied. You must run the command for each component separately, but for the switch, the command lists or upgrades the firmware versions for all network switch components.

Check the Action column to see if any firmware needs upgrading.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Important Do not shut down or reboot your system or any system component until step 6. Special instructions might apply. See step 6 for details. If action is required, disable failover using the StoreOnce CLI command, system disable failover. Before upgrading firmware, stop service sets. serviceset stop all to stop service sets and ensure that no jobs are running Wait one minute and check that service sets have stopped. serviceset show status to verify that all service sets have stopped. It might take up to one minute for all service sets to stop. # serviceset show status Service Set 1 Status ------------- ------Overall: Not Running Service Set 2 Status ------------- ------Overall: Not Running Note The above example is from a StoreOnce System with two service sets—a single-couplet cluster.

Figure 8-25 Multinode systems Use the hardware update firmware command, as required. For example, there might be instances where replacement hardware might need a firmware downgrade to bring it in

||||||||||||||||||||

||||||||||||||||||||

line with the supported firmware version of the installed StoreOnce software on the system being maintained. (Use the force parameter in the command to enable this.) You must specify the node, storage, or switch and repeat the command for each node, storage, and switch in your system. The command syntax is: hardware update firmware <nodeX|storageX|switchX> Important Update the firmware on all nodes, storage components, and switches before shutting down or rebooting the system. Please refer to Customer Notice c04682361 for full details on performing firmware upgrades on StoreOnce Systems.

Note Nodes, storage components, and switches must be updated separately; updating all components simultaneously is not an option from the StoreOnce CLI. In some cases, when the firmware update has completed, you will need to perform a reboot to complete the firmware update. A message is displayed in the CLI output after successful completion, if a reboot is required. The message will also advise what type of reboot is required. Reboot the system using the StoreOnce CLI command: system reboot Power down the nodes using the StoreOnce CLI command: system shutdown After the system has completely shut down, wait at least one minute before powering on using the Power On button on the server(s). Enable failover using the StoreOnce CLI command, system enable failover

Upgrading StoreOnce software To establish which StoreOnce software version is installed, click on StoreOnce in the Navigator and look at the Software Revision under System Information. Check HPE Support to find out if a later software release is available at: www.hpe.com/support/softwaredepot If it is, download the software release and follow the instructions in the accompanying Technet24

||||||||||||||||||||

||||||||||||||||||||

Release Notes to install the software. In a Multinode system, all nodes are updated in parallel. A software update is NOT an online process. Services will shut down to perform the updates, so the whole appliance will be offline for a period of time (approximately 30 minutes). Software updates can only be performed through the StoreOnce CLI using the Admin user account. Software updates are delivered to customers in the form of a single rpm package, which might contain multiple rpms. Everything is updated in the same way, using the rpm package: the software, third-party components, and drivers. Important In a Multinode system, a software upgrade process cannot be performed on a cluster if one or more nodes are down or if any hardware issues are reported. In this case, the node needs to be brought up and working before the software update process. Similarly, if a service set is faulty or if any software issues are reported, the upgrade will not proceed. Example of Gen3 Upgrade process: Back up data as this is destructive. Remove USB key from D2D system. Update Bios and Smart controller to latest firmware. Download Quick restore Image and USB key utility to PC. Insert blank FAT32-formatted USB key and run USB utility. Insert key in D2D system and power on Type “R” to reimage.

Download Quick Restore ISO Image The StoreOnce Quick Restore ISO Image (QR ISO) is used to install the Gen3 software image onto the StoreOnce appliance disks. Go to the HPE Support page at http://www.hpe.com/support/hpesc, search on your product, for example StoreOnce 5500 System, and select HPE StoreOnce Backup. Under Download Options, select the Get drivers, software & firmware link.

||||||||||||||||||||

||||||||||||||||||||

Select OS Independent under Operating Systems to display a list of all available downloads. The QR ISO image is under Software. Click Obtain Software and select the QR ISO image appropriate to your product. You will be required to sign in with HPE Passport to download the QR ISO image and can create an HPE Passport account when you first access the site. Create the physical QR media. Use the HPE USB Key Utility for Windows to copy the ISO image to a USB key using a Windows PC. The HPE USB Key Utility for Windows can be downloaded from the downloads page. It is under Utility. Alternatively, use the integrated Lights Out (iLO) to mount the virtual CD-ROM over Virtual Media. Important HPE recommends that you label the new media for ease of identification.

Create a bootable QR USB Stick Use the HPE USB Key Utility for Windows to copy the ISO image to a USB key using a Windows PC. Note USB 3 keys are not supported. Go to http://www.hpe.com. Click the Support link in the upper right corner to take you to the Support Center site for HPE products. Click the HPE Servers, Storage, and Networking link (under Product Support). As product name, enter HPE USB Key Utility for Windows and click Go. Click drivers, software & firmware (under Download options). Select your product’s operating system and select any Windows version, for example, Windows Server 2012.

Technet24

||||||||||||||||||||

Click the + next to Utility—Tools Click the Download button to download the latest version. Download the Smart Component to a directory on your hard drive and change to that directory. The downloaded file is a self-extracting executable with a filename based on the Smart Component Number. From that drive and directory, execute the downloaded file. After installation, the utility will place a shortcut in HPE System Tools in the Programs start menu folder. To make your drive key bootable and copy the ISO image to it: –

Place your HPE USB drive key (minimum size 4 GB flash stick) in an available USB port.

–

Select the HPE USB Key Utility shortcut in the HPE System Tools folder.

–

Complete each step presented by the application.

||||||||||||||||||||

Once complete, safely remove the USB stick from the Windows PC.

Running a system self-test The StoreOnce CLI command, system confidencecheck, runs a self-test on the system and produces a report of any problems. It is a good practice to run this command before and after maintenance activities. Refer to the HPE StoreOnce System CLI Reference Guide for more information about using this command. system/confidencechecker# runtest all

||||||||||||||||||||

||||||||||||||||||||

Learning check What is a good practice to perform before upgrading BIOS?

Technet24

||||||||||||||||||||

||||||||||||||||||||

Optimizing HPE StoreOnce Storage systems The StoreOnce System is designed for high-performance backup and restores operations according to the specification documented; however, it requires careful configuration. Use the internal reporting facility to assist with optimizing, such as reporting shown in Figure 8-26.

Figure 8-26 Reporting in HPE StoreOnce GUI There are many reasons why you might not see the performance you expected from the StoreOnce system. The network speed or disk access is not fast enough to allow streaming of data. The backup application is not optimized. For example, there are not enough backup streams configured to achieve optimum performance. The StoreOnce System is not optimized for the network and application. Where possible, send the same data types to the same device configured on the StoreOnce system; this will maximize deduplication ratios. Run multiple backups in parallel to improve aggregate throughput to a StoreOnce appliance, using the maximum number of simultaneous streams per configured device. Create separate windows for backup, replication, housekeeping, and offload to physical tape so that overall performance is much more predictable than when all processes interact with each other in overlapping windows. Configure multiple ports in a network bond to achieve maximum available network throughput. Identify other performance bottlenecks in your backup environment such as slow clients and media servers. Best throughput is achieved with multiple streams; the actual number per

||||||||||||||||||||

||||||||||||||||||||

device/appliance varies by model. Choose the backup device type that best fits your needs and is supported by your backup software provider; VTL (virtual tape library), network-attached storage (NAS) using CIFS (Common Internet File System) and NFS (Network File System), and StoreOnce Catalyst device types are available.

Optimizing StoreOnce Catalyst best practices HPE StoreOnce Catalyst is a unique interface and is fundamentally different from virtual tape or NAS. It provides the backup application with full control of backup and replication (called Catalyst Copy). For this reason, best practices are dependent on the backup application. Ensure that the media servers where Catalyst low bandwidth backup is to be deployed are sized accordingly; otherwise, the implementation will not work well. As with other device types, the best deduplication ratios are achieved when similar data types are sent to the same device. Best throughput is achieved with multiple streams; the actual number per device/appliance varies by model. Because Catalyst stores can be acting as a backup target and inbound replication target, the maximum value applies to the two target types combined (although inbound copy jobs would not normally run at the same time as backups). Although Catalyst copy is controlled by the backup software, the copy blackout window overrides the backup software scheduling. Check for conflicts. The first Catalyst low bandwidth backup will take longer than subsequent low bandwidth backups because a seeding process has to take place. If you are implementing multihop or one-to-many Catalyst copies, remember that these copies happen serially not in parallel. Ensure the backup clean-up scripts that regularly check for expired Catalyst Items run at a frequency that avoids using excessive storage to hold expired backups (every 24 hours is recommended). There are several specific tuning parameters dependent on backup application implementation.

Optimizing VTL best practices Make use of multiple network or Fibre Channel ports throughout your storage network to eliminate bottlenecks. Technet24

||||||||||||||||||||

||||||||||||||||||||

For FC configurations, split virtual tape libraries and drives across multiple FC ports (FC VTL is available on some models, such as StoreOnce B6200, 4210 FC/4220/4420/4430 models). Configure multiple VTLs and separate data types across them, for example, SQL to VTL1, File to VTL2, and so on. Configure larger “block sizes” within the backup application to improve performance. Disable any multiplexing configuration within the backup application. Disable any compression or encryption of data before it is sent to the StoreOnce appliance. Best throughput is achieved with multiple streams; the actual number per device/appliance varies by model. Schedule physical tape offload/copy operations outside of other backup, replication, or housekeeping activities.

Optimizing NAS best practices Configure multiple shares and separate data types into their own shares. Adhere to the suggested maximum number of concurrent operations per share/appliance. Choose backup application container file sizes so that the entire backup job fits within that container size. For example, if a full backup is 500 GB, set the container size to be at least 500 GB. Each NAS share has a 25,000 file limit; some backup applications create large numbers of small control files during backup to disk. If this is the case, it might be necessary to create additional shares and distribute the backup across multiple shares. Disable software compression, deduplication, and synthetic full backups. Do not preallocate disk space for backup files within the backup application. Best throughput is achieved with multiple streams; the actual number per device/appliance varies by model. For NFS shares ensure that the correct mount options are used to ensure in-order delivery and provide better deduplication ratios. See the “HPE StoreOnce Linux and UNIX Configuration Guide” for specific details.

Optimization considerations with deduplication on

||||||||||||||||||||

||||||||||||||||||||

StoreOnce The inline nature of the deduplication process means that it is a very processor and memory intensive task. HPE StoreOnce appliances have been designed with appropriate processing power and memory to minimize the backup performance impact of deduplication. Best performance will be obtained by configuring a larger number of libraries/shares/Catalyst stores with multiple backup streams to each device, although this has a trade off with overall deduplication ratio. If servers with lots of similar data are to be backed up, a higher deduplication ratio can be achieved by backing them all up to the same library/share/Catalyst store, even if this means directing different media servers to the same data type device configured on the StoreOnce appliance. If servers contain dissimilar data types, the best deduplication ratio/performance compromise will be achieved by grouping servers with similar data types together into their own dedicated libraries/shares/Catalyst stores.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Optimizing general best practices: HPE 3PAR StoreServ 8000 initial configuration System considerations include: The minimum supported raw capacity is equal to six-min drive size available. The storage base must have either zero or an even number of drives of the same type. An all-LFF configuration is supported. The storage base can be deployed with no drives as long as the attached LFF enclosures contain the minimum required quantity of drives. Four‐node configurations should be balanced. A best practice is to identically configure both node pairs. At minimum, deploy the same types of drives on both node pairs, even if the drive quantities differ between node pairs. When the configuration includes multiple enclosures, the best practice is to use RAID configurations that permit an HA cage. The HA cage configures the RAID striping so that the system can withstand the loss of an entire drive enclosure. The HPE 3PAR StoreServ x4x0 two‐node storage base includes two 1U rack filler panels. The best practice is to install filler panels in the 2U space directly above the 2U node enclosure, keeping that space empty in case the user wants to upgrade the array. Enclosure considerations include: Minimum enclosure configuration is a storage base with no additional drive enclosures. The drive enclosures must have an even number of drives, with a minimum of two. Empty drive enclosures are not allowed as they must have 2 drives unless it is a base enclosure, and you have an LFF enclosure with 8 drives attached. NL, FC, and SSDs can be mixed in the same enclosure. When configuring an array that includes LFF and SFF drive enclosures, arrange the enclosures in the rack so that all the 2U enclosures (drive and node enclosures) belonging to one node pair are together, and all the 4U enclosures belonging to the node

||||||||||||||||||||

||||||||||||||||||||

pair are together. Drive considerations include: Minimum drive quantity per node pair is 6* for FC and SSD (8 recommended), and 12 for SAS Nearline. The minimum applies to each new drive type included in the array. HPE strongly recommends RAID6 for FC and NL drives. To enable encryption, all the drives in the system have to be SEDs. No partially encrypted systems. You can have a mix of SEDs and non-SEDs in a system if not using encryption. SSD-specific configuration practices. SSD‐only configurations are permitted. A maximum of 120 SSDs per node pair is allowed. (240 SSDs in 8448/8450)

HA disk: Two-node drive upgrade best practice Lighter drives, shown in Figure 8-27, are the originals and darker drives are the upgrade drives. This is a 2-node x200 system with 1 enclosure running R1 and R5 3+1 with 8 installed drives.

Figure 8-27 HA disk: Two-node drive upgrade best practice In this instance, 8 drives or an increment in multiples of 8 drives would be the best practice, depending on the capacity needed go to 16 or 24 drives. Eight disks are recommended, so each of 2 nodes would have a full set size (2x[3+1]). For SSZ=5, 10 disks would be recommended. You will be able to create new logical disk (LD) space as needed and, in the end, have the best capacity utilization possible with no performance implications.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HA disk: Two-node enclosure upgrade best practice Lighter components, shown in Figure 8-28, are the originals and darker components are the upgrade components. This is a 2-node x200 system with 1 enclosure running R1 and R5 3+1 with 24 installed drives.

Figure 8-28 HA disk: Two-node enclosure upgrade best practice In this instance, 8 drives or an increment in multiples of 8 drives would be the best practice, depending on the capacity needed go to 32 or 40 drives plus the additional enclosure. You will be able to create new LD space as needed and, in the end, have the best capacity utilization possible with no performance implications.

3PAR OS 3.3.1 and active-active disks 3PAR OS 3.3.1 introduced the concept of Active-Active Disks for all types of physical disks (PDs). All PDs between a node pair are included in the pool of PDs used to make an LD for either node. This means that LDs can now be formed on each node of a node pair with half the number of disks previously needed. For example, eight disks were recommended, so each of two nodes would have a full R5 set size (2x[3+1]). Since 3.3.1, you will be fine with four as both nodes can share same disks: 3+1 set size. 3PAR OS downgrade warning: If, for example, you are using 6 drives for R6 (4+2) and you downgrade below 3PAR OS 3.3.1, then you will need 12 drives (2x[4+2]). The existing LDs will continue to function. However, new RAID6 LDs cannot be created after the downgrade because of PDs only being owned by a single node. The common provisioning group (CPG)

||||||||||||||||||||

||||||||||||||||||||

settings will need to be adjusted to use RAID5 or more PDs will be needed after the downgrade. With 3.3.1, HPE is now recommending RAID6 for all volumes (default set size is 8, minimum set size is 6) and recommends that the RAID set size is less than the number of disks attached to the node pair. For example: With RAID5 set size 4, a minimum of 6 disks is recommended. With RAID6 set size 8, a minimum of 10 disks is recommended. The reason for this is that when “number of disks == RAID set size,” any disk failure will result in degraded RAID sets and no new RAID sets will be able to be formed, causing virtual volume (VV) growth failures and stale snapshots.

How set size affects rebalancing activities and upgrades There is always a minimum and a best practice. Everything between the minimum and the best practice is fully supported. Whether or not it is a good idea or will impact the customer in a negative way is up to the Solution Architect to determine. If the enclosure count per node pair matches the set size and HA Enclosure/Cage is used but you are adding another enclosure and leaving the set size alone, the drive count does not have to be an increment of the set size. There is a set size enforcement option in NinjaSTARS under the options that can be turned on or off. There are two scenarios where it is recommended to leave it on. The first is when the enclosure count per node pair matches the set size and HA Enclosure is used. The second is when the system is routinely ran close to 100% capacity utilization and then a rebalance is attempted.

Mixed-drive capacities and speeds When you have mixed-drive capacities in the same tier sharing the same CPG characteristics, you have two choices: Stay with one CPG: until the smaller drives are consumed from a capacity standpoint, you will still have the widest stripe available for performance and then the stripe gets

Technet24

||||||||||||||||||||

||||||||||||||||||||

narrower. Create a second CPG: with a disk filter for the larger drives only. For volumes that have lower performance requirements, use the CPG with the disk filter saving the wider stripe for the volumes that have high-performance requirements. This does require an extra decision point when provisioning volumes. If mixing drives in the same tier that have different speeds, such as 10K and 15K SAS drives, you should assume that a volume will get the 10K performance. When mixing different SSDs, be aware that the different capacities can have more impact because of the smaller drive counts and the impact to sparing levels. When mixing different SSDs that are MLC (multi-level cell) and cMLC (consumer grade MLC), be aware of the delta in write performance between these drive types and how it could affect the existing performance of volumes currently only on MLC drives. It might be better to stay with all MLC drives or have separate CPGs depending on the application workloads being supported. The higher the performance needs and write ratio of the application, the more important this becomes. If mixing drives in the same tier that have different performance, such as cMLC and MLC drives, you should assume that a volume will get the lower performance.

All-flash array considerations For four-node systems with SSDs, shown in Figure 8-29, it is especially important that the number and capacities of the SSDs be evenly distributed between node pairs.

||||||||||||||||||||

||||||||||||||||||||

Figure 8-29 All-flash array considerations The number of SSDs and the total capacities of those SSDs should be balanced. A small amount of difference in drive count or capacity can make a large imbalance, which can cause the system to not deliver the full performance of the hardware. Mixing same-type drives of different capacities leads to depletion of extra capacity in creating additional spare space capacity. It is best to use same capacity drives, but different capacity drives can be mixed. If different capacities drives are used, best practice is to add these in increments of the set size per node pair. Because of the increased cost per GB of SSD space, it is best to make the most utilization of these drives.

Adaptive Flash Cache HPE 3PAR Adaptive Flash Cache is included as part of the HPE 3PAR Operating System Suite version 3.2.1 and later and is supported in all HPE 3PAR StoreServ Storage arrays that have a blend of SSDs and HDDs. Benefits Reduced latency for random read-intensive workloads. Technet24

||||||||||||||||||||

||||||||||||||||||||

Responds instantly, providing smart and flexible second-stage data caching based on application and workload demands. Enables HPE 3PAR Adaptive Flash Cache across the entire system or select particular workloads to accelerate. Requirements HPE 3PAR OS version 3.2.2. Two SSDs in 8000/9000 series or four SSDs in the 20000 series per node pair. Best practice: On systems with four or more SSDs per node pair available, enable flash cache system wide.

Increasing and decreasing the size of Adaptive Flash Cache The configured Adaptive Flash Cache (AFC) size on a system cannot be changed dynamically. To increase or decrease the amount of AFC configured on the system, you must first remove AFC by issuing the removeflashcache command (this disables AFC for all VVs) and then recreate AFC with the createflashcache command specifying the new size. Use the createflashcache command to specify how much SSD capacity on each node pair should be allocated for AFC use. All node pairs will be configured for the same amount of AFC, so all node pairs must have enough available SSD space to satisfy the amount of AFC specified in the createflashcache command. The quantity of AFC to be enabled can be specified either in gigabytes or terabytes. If you specify the amount of AFC in gigabytes, it must be in 16 GB increments. When AFC is created on the array, it is created as a special RAID1 logical disk behind each node in the array that is named fcacheld.x, where x is the node ID. All existing configuration rules created with the setflashcache command will persist upon AFC removal and recreation.

Adaptive Flash Cache simulator For HPE 3PAR StoreServ arrays running HPE 3PAR OS 3.2.1 and later, there is an AFC simulator that can be run even if the system does not contain any SSDs. The simulator allows the HPE 3PAR StoreServ array to track AFC statistics without actually using SSD space. These statistics can then be analyzed by using the standard AFC CLI

||||||||||||||||||||

||||||||||||||||||||

commands to determine whether you should add SSDs to the HPE 3PAR StoreServ array and configure AFC. It can even be used to determine how much AFC to create and configure. Use the –sim option to the createflashcache command to enable the AFC simulator: cli% createflashcache –sim <size> After the simulator has been started, use the setflashcache command to specify the VVs that should be considered (or the entire system) for AFC. After the system has run for some time (hours or days), you can then view the AFC statistics by issuing the statcache command (or the srstatcache command to review historical data) to determine how much benefit can be derived from enabling AFC for the system or for specific VVs or virtual volume sets. The simulator can be easily used to try out different sizes of AFC on a system to help you find the right mix. Remember, the larger the configured AFC, the better the probability for cache hits. A balance needs to be struck between the size of the AFC and how many hits it receives. When you are finished with the simulator and ready to enable AFC in real time, run the createflashcache command and specify the quantity of AFC you need per node pair. All the configuration rules created via the setflashcache command as the simulator is running will be enabled automatically. Note It will take some time for the AFC to warm up after you start it. Remember to stop the simulator by issuing the removeflashcache command even if you have decided not to enable AFC, because the simulator consumes some system resources (CPU and memory) when it is running, and leaving it in place wastes these system resources. Any policies you created by issuing the setflashcache command will be maintained unless you remove them by using the disable option with the setflashcache command. Note An HPE 3PAR System Reporter license is not necessary to display AFC statistics with the srstatcache CLI command.

Balancing AFC after adding SSDs to the system Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR Autonomic Rebalance provided via the tunesys operation does not apply to AFC logical disks. After adding additional SSDs to a system, you can rebalance AFC across the new SSDs by first removing AFC with the removeflashcache command (this disables AFC for all VVs) and then recreating AFC with the createflashcache command. Rules created with the setflashcache command persist when AFC is recreated. Important Because AFC must first be removed to change its size or to rebalance across new SSDs, HPE recommends that you do this during a maintenance period if possible because it will take some time for the AFC to warm up after being recreated. The warmup period might have a noticeable effect on system performance.

HPE 3PAR Dynamic Optimization Dynamic Optimization (DO) is an HPE 3PAR OS feature that enables the administrator to dynamically optimize volumes performance by changing parameters, such as RAID level, set size, and disk drive types. DO allows for the nondisruptive restructure of virtual volumes. This functionality promotes system optimization through improved utilization of all physical resources present at a given time. In addition, DO enables altering of service levels associated with a given volume by allowing the administrator to change the volume parameters. This feature allows the administrator to change service levels when either incorrect application performance information has been given or history shows that the volume is either under- or overusing the resources that have been provisioned. During a peak period when performance is critical, the VV can be optimized by moving to faster disks (NL disks to FC disks or SSD) and the RAID level. Tiering for ILM allows for movement to a less expensive set size, and using less expensive NL disks, resulting in optimizing cost savings. DO is an optional feature that allows entire VV/VLUN (Virtual Volume and LUN pairing) movement from one CPG to another for cost reasons, performance reasons, or space utilization reasons in real time with no downtime. Adaptive Optimization (AO) is different in that data from VV/VLUN can be distributed across various media using subvolume-based algorithms. Differences between AO and

||||||||||||||||||||

||||||||||||||||||||

DO are shown in Figure 8-30.

Figure 8-30 HPE 3PAR Dynamic Optimization

HPE 3PAR Adaptive Optimization AO uses SR statistics gathered from LDs and physical disks to relocate customer data on physical volumes in an optimal way. AO relocation accomplishes two primary goals: Increase performance of frequently accessed regions of data by moving those regions to higher tier storage (for example, moving to SSDs from spinning media). Improve cost-efficiency by moving lightly accessed regions of data to a lower performance and less expensive tier of storage (for example, moving from regular drives to NL drives). Subvolume optimization software allows striping of data across storage tiers. Cost optimization occurs automatically through defined policies in the storage array.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Because storage tiers can be of different RAID types, capacity efficiency is maximized by using only RAID1 for the most frequently accessed storage and by using RAID5 or RAID6 for less frequently accessed storage. Other benefits include: AO can migrate data from a fully occupied tier of storage to a tier that has more available capacity. AO can be regularly scheduled to adjust the data layout as your data usage changes over time. For well-known applications that require a high level of performance, use tailored AO configurations that: Preferably use all the tiers available in the system. Execute immediately at the end of the high-activity period. Use a measurement/hour that covers only the length of the high-activity period. Use a mode of “performance.” For test environments where performance is not a requirement, use an AO configuration that: Uses only FC and NL tiers. Run during the days of the workweek only (for example, Monday–Friday). Executes once a day, preferably at night. Uses a measurement/hours of 24. Use a mode of “balanced.”

HPE 3PAR Priority Optimization As of HPE 3PAR OS 3.1.2 MU2, Priority Optimization QoS (quality of service) offers the ability to set a maximum limit of IOPS or bandwidth available to a VVset (virtual volume set). Since HPE 3PAR OS 3.1.3, this feature set was extended to add priority levels (high, normal, and low), latency goals, and minimum goals along with the ability to set QoS rules against Virtual Domains. Care must be taken when setting QoS rules, system performance has to be understood and historical performance data has to be taken into account when setting maximum limits, minimum goals, and latency goals.

||||||||||||||||||||

||||||||||||||||||||

Realistic latency goals for NL >= 20ms, for FC >=10ms, and for SSD >=1ms.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check The Adaptive Flash Cache simulator can only be run in the CLI. True False

||||||||||||||||||||

||||||||||||||||||||

Tuning HPE 3PAR StoreServ systems All upgrades should be performed with the aim (as far as possible) of balancing the PD devtypes (SSD, FC, and NL) and capacities between node pairs. It is possible to install an unbalanced configuration with more of some devtypes on some node pairs compared with others. In this case, tunesys will distribute data in the same way as new volumes are laid out and use (as much as possible) an equal amount of space on each node regardless of the number of PDs of a particular devtype behind each node pair.

Introduction to tunesys HPE 3PAR System Tuner Software, shown in Figure 8-31, spreads volumes evenly and widely across all available resources to deliver balanced performance and autonomically and nondisruptively detects potential bottlenecks and hot spots. The software rebalances volumes to maintain peak performance without disrupting applications, impacting service levels, or changing pre-existing service-level characteristics.

Figure 8-31 Introduction to tunesys

Technet24

||||||||||||||||||||

||||||||||||||||||||

Tunesys is the primary command in the HPE 3PAR OS for redistributing data to new storage capacity after a system has been upgraded. Tunesys should only be run after a hardware upgrade is completed; it does not require a license. The command performs a point-in-time analysis of different aspects of the layout of VVs and LDs and determines which VVs, LDs, or PDs need to have their layout and space distribution modified to best use the new capacity. Tunesys then acts as a task controller—starting and monitoring individual tuning subtasks to effect the rebalance. CPGs involved in tuning will be compacted periodically. After each tune, tunesys waits for chunklet cleaning to maximize the available layout for the next tune. Different types of low-level tunes are started to balance the system in different ways depending on the type of upgrade detected. There are 3 main types of upgrades: Node pairs are added (Phase 1 Internode tuning). New Cages or PDs are added to existing nodes (Phase 2 Intranode tuning). Where the LD characteristics differ from the parent CPG (Phase 3 Logical disk relayout tuning).

Tunesys phases The tunesys command runs between one and three phases, depending on the options specified and what optimizations are required on the 3PAR StoreServ: Phase 1 uses a CLI command tunevv to create new Logical Disks using the parameters inherited from the parent CPG. Phase 2 analyzes the disk utilization across all the disks that belong to each node. Moves data from overallocated disks to other disks to try to balance utilization. Phase 3 detects when the characteristics of logical disks that belong to CPGs do not match the characteristic of the parent CPG.

Phase 1—Internode tunes When new node pairs are added to a storage system, existing VVs can use the new hardware if they are retuned. The underlying logical disks must be redistributed across new disks or new logical

||||||||||||||||||||

||||||||||||||||||||

disks. VVs that have poor balance between nodes are retuned, so they are better spread between nodes. Use the -nodepct (% difference between nodes) to specify how much of an imbalance between node pairs can be tolerated before tuning is run. A warning is generated if the percentage difference in PD counts between node pairs is exceeded (controlled by the spindlepct). This can be overridden.

Phase 2—Intranode tunes Analyzes the disk utilization across the disks that belong to each node. Moves data from overallocated disks to other disks to try to balance utilization. HPE 3PAR OS 3.2.1+ uses a new algorithm—A mixture of chunklet movement and (blockless) LD tuning. Faster chunklet movement. Balances better, but at the expense of more data movement and a longer run time. Analysis and movement are performed by device type: NL, FC, and SSD. You can also decide to tune the specified device type only with -devtype option. This phase is used only when an imbalance occurs in space utilization across PDs behind a node.

Phase 3—Logical disk relayout tunes Detects when the characteristics of logical disks that belong to CPGs do not match the characteristic of the parent CPG. This can happen if the CPG characteristics have been changed—for example, the setcpg command can be used to change the RAID type or set size of a CPG. This change only affects newly created VVs and logical disks. Existing ones are not modified until a tune is done. Logical disks that do not match are recreated using the current CPG characteristics. The data is moved to the new logical disks, and the old logical disk is removed (if possible). Can relayout an LD to more or fewer PDs.

Technet24

||||||||||||||||||||

||||||||||||||||||||

All upgrades should be performed with the aim (as far as possible) of balancing the PD devtypes (SSD, FC, and NL) and capacities between node pairs. Note that on HPE 3PAR StoreServ 7200 and 8200 series systems, tunesys will run automatically after the admithw command completes. Note The admithw command admits new hardware into the system. If new disks are discovered on HPE StoreServ 7200/c and 8200 systems, tunesys will be started automatically to redistribute existing volumes to use the new capacity unless the -notune option is used. On other systems, tunesys needs to be run manually after hardware installation. Also as admithw runs automatically (after a short delay) after PD insertion on 7xxx systems, PD insertion on these systems will always trigger tunesys. Tuning has a lower priority than serving data and the same priority as other tasks; therefore, it can take a long time to run.

Tunesys Tunesys runs as a background task and can take a very long time. The exact duration is difficult to predict and depends on a number of factors including: How much new capacity has been added How full the existing PDs are (utilization level) The type and speed of the new and existing PDs The I/O load on the system Tuning is regarded as a background activity and will try to place a minimum additional load on a system Customer I/O always takes priority. When an LD is deleted or trimmed (via compaction), some or all of the chunklets belonging to the removed regions are freed back to the system for reuse (Chunklets are freed in multiples of LD “rows” trimmed from the end of LDs after compaction). Before they can be used again, these chunklets need to be initialized. This process takes a certain amount of time for each chunklet as the disk surface has to be physically accessed for each block of data. This typically takes place at a rate of a few chunklets per PD per minute (the rate varies depending on the speed of the PD and I/O load on the

||||||||||||||||||||

||||||||||||||||||||

PD). To maximize the I/O performance of the system, wide striping is used; this relies on maximizing parallel accesses to PDs and uses an optimal number of PDs in each LD. When an LD is created or grown, it is allocated free space from the pool of clean chunklets. Ideally, the underlying algorithm will create RAID sets and form LD rows from a large number of PDs. If some PDs do not have enough clean and free chunklets to fulfill the growth or creation request, then they will not be used for that LD, potentially reducing performance and placing more I/O load on other PDs. When a tune is started, destination regions are formed by either creating new LDs or growing existing ones. If a reduced pool of PDs is available (because of lack of clean space), the new LDs will have a reduced stripe size and therefore some reduction in performance. New tunes are not started until compaction and cleaning is complete, and waiting for these operations to finish can add a considerable amount of time to a tunesys run.

Scheduling tunesys Tunesys can be run at regular intervals using the standard OS task scheduler. If a previous run has not completed, a newly started instance will exit immediately and will not affect the running instance. Example schedules: createsched “” “” <schedule name> createsched “tunesys –f” @weekly tunesys_sched Care should be taken with scheduling tunesys if any class of drive is projected to be close to its maximum IOP capacity during the estimated tuning period. There is little penalty to canceling tunesys at peak times and restarting it again during off-peak periods. Given how long tunesys will run after a large upgrade, the most useful schedules are @weekly and @monthly.

Automatic tunesys In OS version 3.2.2, tunesys starts automatically on some system after admithw completes. This applies only to HPE 3PAR StoreServ 7200x and 8200 systems: These systems have a fixed number of nodes (2) and are not node-pair upgradable, so Technet24

||||||||||||||||||||

||||||||||||||||||||

the number of use cases is limited. The default for these fixed-node-configuration, two-node systems is to autotune, but if a complicated upgrade (or staged upgrade) is being performed (such as adding cages and PDs over a number of days), HPE recommends that you perform a single tune at the end of the upgrade. Automatic tuning will not start if the system has failed or degraded drives or has a spindlepct warning message. Tuning must be started manually in these circumstances. Note that 7200x systems start admithw automatically when new disks are inserted. This will start tunesys automatically without user intervention. If you do not want to run tunesys after every disk insertion, please cancel the tunesys task each time. When to use: When new drives or cages are added to existing nodes. When new nodes pair with an equivalent number of drives as existing ones are added. Important to know Understand which phase will be used before you run the command. The tunevv and tuneld commands use additional space during the region move operation; hence, usable space will reduce. Does not need a DO license. Starting tunesys part way through an upgrade will do no harm but is inefficient and will take longer and move more data than a single run at the end.

Tunesys considerations Run tunesys after hardware upgrades on nodes, cages, magazines, and disks. Until tunesys is run, the new hardware can only be used by newly allocated capacity. This can lead to an imbalance in the 3PAR. No user interaction is needed after it starts. Tuning has a lower priority than serving data and the same priority as other tasks; therefore, it can take a long time to run. Allow a dry run of tunesys to see which type of tuning will be suggested for the current configuration (limited to internode tuning). The Tune System feature allows you optimize system performance by reallocating space for virtual volumes and logical disks and reallocating chunklets across physical disks.

||||||||||||||||||||

||||||||||||||||||||

Detects physical disk space usage imbalance between nodes and rebalances virtual volumes between nodes. Detects newly added disks and rebalances chunklets from a node’s existing physical disks to its underallocated ones. Performs a relayout of LDs whose characteristics do not match their parent CPG. Note Tuning a system can take anywhere from a few hours to several days. If the system has restarted during that time, you might not see the task information, and you should check the log for failures.

SSMC—Accessing Tuning from SSMC In the StoreServ Management Console (SSMC), open the Mega menu and go to Storage Systems → Systems. Select your system on the left, and then from Actions menu on the right, select Tune, as shown in Figure 8-32.

Figure 8-32 SSMC—Accessing Tune The SSMC supports an Analyze only option; this allows the Tune to run to completion identifying any phases that require action but nothing is moved, as shown in Figure 8-33. Change the Analyze option to Yes (optional, for preview-only run) and press the Tune Technet24

||||||||||||||||||||

||||||||||||||||||||

button.

Figure 8-33 SSMC—Tuning Click the activity icon → System Tuning activity → Details to see the progress, as shown in Figure 8-34 and Figure Figure 8-35.

Figure 8-34 SSMC—Tuning progress

||||||||||||||||||||

||||||||||||||||||||

Figure 8-35 SSMC—Tuning progress Expand the task and click Task Detail to see more information, as shown in Figure 8-36.

Figure 8-36 SSMC—Tuning progress

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Learning check Tunesys phase 3 detects if new disks were added and will automatically redistribute the data across them. True False

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ operating system upgrade The OS Upgrade must be performed by an HPE support personnel in some regions and/or for some products and/or for some OS versions. Before you begin with 3PAR OS upgrade, you should always check the respective “HPE 3PAR Operating System Upgrade Planning Guide.” It includes information about: An overview of the online and offline upgrade processes for the HPE 3PAR OS. Prerequisites for online upgrades. Guidelines for online upgrades. Planning information needed for online upgrades. Veritas Dynamic Multipathing prerequisites for online upgrades. Remote vs. local methods of upgrading. The HPE 3PAR StoreServ Storage supports two different methods for upgrading the HPE 3PAR OS: online and offline. All online upgrades are node by node. Each array node is upgraded one at a time. A simple or auto node-by-node upgrade occurs with each node upgraded in an HPE 3PAR OS predetermined sequence and timing. Offline is used when the system is operating several levels below the upgrade package version. All nodes are upgraded and then restarted at the same time. The system is unavailable to the hosts during this operation. The upgrades can be performed with a service representative on-site or remotely (from HPE). Remote upgrades require that remote operations are available. See the HPE Storage Single Point of Connectivity Knowledge (SPOCK) at http://www.hp.com/storage/spock for HPE 3PAR support matrixes for HPE 3PAR OS versions supported by each array model. HPE recommends updating host drivers to a supported level of the targeted HPE 3PAR OS upgrade version before attempting an online upgrade. For this information, see the HPE SPOCK website. These documents are provided to help you set up hosts with HPE-supported configuration information.

||||||||||||||||||||

Note In addition to this planning document, see the HPE 3PAR host OS implementation guides available from the HPE Storage Information Library at http://h17007.www1.hpe.com/us/en/storage/info-library/index.aspx

Planning information for online HPE 3PAR OS upgrades The host connectivity environment must be supported. HPE service representatives will request this information, as shown in Figure 8-37, for review before performing an online upgrade. Relevant details include: Host platform Architecture (for example, SPARC/x86) Operating system (version and patch level) Host multipath software HBA and CNA Switch Information on host software provided by HPE, examples include: –

HPE 3PAR Recovery Manager (SQL, Exchange, Oracle)

–

HPE 3PAR VSS Provider

–

HPE 3PAR MPIO for Microsoft Windows

–

HPE 3PAR ODM Definition for IBM AIX

–

HPE 3PAR System Reporter

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 8-37 Planning information for online HPE 3PAR OS upgrades In addition to verifying that the driver and firmware levels match the information in HPE SPOCK, host configuration settings might be needed as documented in the HPE 3PAR host OS implementation guides. For more information, see the interoperability information on the HPE SPOCK website. Verify these settings before performing an online upgrade: All attached hosts must have multipathing software configured and operational for every volume exported. A host definition must not contain initiators from multiple hosts (separate for each cluster member). iSCSI hosts must have active I/O on all paths during the upgrade. The recovery manager for Oracle, SQL, Exchange, or VMware might require a newer version to be compatible to the new version of the HPE 3PAR OS. Verify compatibility of any host third-party software to the HPE 3PAR OS release. Always verify whether SSMC or the CLI requires an upgrade to be compatible with the new version of the 3PAR OS.

||||||||||||||||||||

||||||||||||||||||||

Always check the HPE 3PAR Operating System Upgrade Planning Guide for special notes on Remote Copy (RC), CLX, Peer Persistence (including Quorum Witness), guidelines for respective operating systems, and possible Host Persona changes. A host definition must not contain initiators from multiple hosts. For example, host cluster configurations must use a separate host definition for each host in the host cluster. iSCSI hosts must have active I/O on all paths during the upgrade. The upgrade process restarts nodes and checks that all hosts are reconnected after each node restarts. The upgrade will not advance until all hosts re-establish connection. An idle iSCSI host will not reconnect; it is important to have I/O on all paths during the upgrade. The HPE 3PAR Recovery Manager software for VMware vCenter requires upgrade to HPE StoreOnce Recovery Manager Central for VMware vCenter v1.1 to be compatible with HPE 3PAR OS 3.2.x. The upgrade should be completed on multiple sites if the Recovery Manager configuration is an RC multisite configuration. The HPE 3PAR Management Console is generally compatible with all HPE 3PAR OS 3.1.1, 3.1.2, and 3.1.3 MU1 versions. For the latest features and fixes, it is advisable to upgrade to the current HPE 3PAR Management Console release. Verify compatibility of any host third-party software to the HPE 3PAR OS release. When you upgrade the HPE 3PAR StoreServ Storage to HPE 3PAR OS 3.1.x, upgrade the HPE 3PAR System Reporter application compatible with 3.1.x with the bundled HPE 3PAR System Reporter CLI Remote Client. (See the HPE SPOCK website.) The HPE 3PAR Management Console is generally compatible with all HPE 3PAR OS 3.1.1, 3.1.2, and 3.1.3 MU1 versions. For the latest features and fixes, you should upgrade to the current HPE 3PAR Management Console release. Always check the HPE 3PAR Operating System Upgrade Pre-Planning Guide for special notes on RC, CLX, Peer Persistence, and possible Host Persona changes. Also verify guidelines for respective operating systems. You can find them in the HPE 3PAR Operating System Upgrade Pre-Planning Guide.

HPE 3PAR StoreServ Storage configuration guidelines Guidelines include: Technet24

||||||||||||||||||||

||||||||||||||||||||

The HPE 3PAR StoreServ Storage must be in a “healthy” state. Automated check is performed. A VLUN for each exported volume must be exported to a defined host on at least two fully operational paths that span adjacent storage server nodes (0 & 1, 2 & 3, 4 & 5, and 6 & 7) in order for HPE 3PAR Persistent Ports to work. HPE 3PAR software applications running on the HPE 3PAR StoreServ Storage must be shut down. Ensure that no HPE 3PAR Virtual Copy promotes are in progress. Verify all the HPE 3PAR RC guidelines in the HPE 3PAR Operating System Upgrade Pre-Planning Guide. The HPE 3PAR StoreServ Storage must be in a “healthy” state (with no failed or degraded physical disks, cages, FCALs, nodes, and so on). The HPE 3PAR OS upgrade software performs an automated check and, if the system is not in a healthy condition, causes the upgrade to fail. A VLUN for each exported volume must be exported to a defined host on at least two fully operational paths that span adjacent storage server nodes, an even-numbered node followed by a consecutive odd-numbered node, on the HPE 3PAR StoreServ Storage. (Valid pairings are 0 & 1, 2 & 3, 4 & 5, and 6 & 7.) HPE 3PAR software applications running on the HPE 3PAR StoreServ Storage must be shut down. Ensure that no HPE 3PAR Virtual Copy promotes are in progress. If any promotes are in progress, wait until the promote finishes to begin the online upgrade. Verify all the HPE 3PAR RC guidelines in the HPE 3PAR Operating System Upgrade Pre-Planning Guide.

Preinstallation requirements Stop all system administration activities while performing an HPE 3PAR OS online update. Verify the Service Processor Software level. Always verify supported HPE 3PAR OS level update paths. When upgrading from 3PAR OS 3.1.1 x to 3PAR OS 3.1.3 MU1, a multistep upgrade is

||||||||||||||||||||

||||||||||||||||||||

required. The first step is to upgrade to 3PAR OS 3.1.2 GA, and then the second step is to upgrade to 3PAR OS 3.1.3 MU1. Additional examples of upgrade paths are shown in Figure 8-38.

Figure 8-38 Upgrade paths When upgrading from 3.1.1 GA to 3.1.1 MU2, there is an issue that requires an offline upgrade. If it is not permissible to keep RC groups online for an upgrade, the RC groups must be stopped before upgrade.

Preupdate tasks Verify the HPE 3PAR OS software level and health of the storage system: The current HPE 3PAR OS level must meet the supported update path requirements.

Technet24

||||||||||||||||||||

||||||||||||||||||||

You need to verify the HPE 3PAR OS level currently running on the storage system. Before starting the HPE 3PAR OS update, determine the current status and verify the overall health of the system. For step-by-step installation procedures, please consult the “HPE 3PAR Customer Self Update Instructions Guide.” To verify the HPE 3PAR OS level currently running on the storage system: Start a CLI session from the SPMAINT main menu, enter =7 and then press Enter. Select the desired storage system (StoreServ). To display the HPE 3PAR OS version, patch information, and the version numbers of other system components, enter showversion -a –b. Enter exit and then press Enter to return to the SPMAINT main menu. To start a health check: From the SPMAINT main menu, enter: =4.4 and then press Enter. Enter the number corresponding to the storage system (StoreServ) you want to run the Health Check on and press Enter. Enter y to retrieve and transfer the check health data and press Enter. After the health check completes gathering the data, the Service Processor (SP) displays a list of files to view. Review and correct any exceptions before performing an HPE 3PAR OS update.

HPE 3PAR OS upgrade methods (3PAR OS pre-3.3.1) There are three different methods by which the array can be upgraded: SP GUI/Service Processor Onsite Customer Care (SPOCC) SP service menu HPE 3PAR array CLI Those methods change after the system is moved to 3PAR OS 3.3.1 and SP 5.0. SPOCC: The easiest and simplest way to upgrade the array is using the SPOCC GUI within the SP. Both the SP and the HPE 3PAR array can be updated by using the GUI. Each of the above choices has different menu-driven screens, which guide the user through the steps

||||||||||||||||||||

||||||||||||||||||||

needed to update either unit. SP service menu: The SP service menu is available to advanced users and can be used to accomplish many different tasks not related to upgrades. The use of this option however should be limited to experienced users on the array such as field service personnel. Features for purposes such as to upgrades are not fully documented as with the use of the SPI GUI and are recommended for HPE service personnel. 3PAR CLI: This method for upgrades does not upgrade the SP and is only used to upgrade the array components. The software to upgrade must be prestaged on the array nodes before the upgrade process. This method is limited in scope and should only be completed by HPE Support personnel only.

HPE 3PAR OS upgrade methods (3PAR OS 3.3.1+ and SP 5.0+) There are two different methods by which the array can be upgraded: SP Service Console (SPSC/GUI) HPE 3PAR array CLI This functionality is only usable when upgrading from 3.3.1, to include patches and 3.3.1.MU1. All upgrades should be driven by the SP. CLI interfaces are still available if needed (mainly for HPE). 3PAR OS 3.3.1 upgrade: Customers cannot upgrade their storage systems to 3PAR OS 3.3.1. Please contact a support engineer for any additional information needed. To arrange for HPE to perform the upgrade, contact the Hewlett Packard Enterprise global deployment center at [email protected] and include the 3PAR StoreServ Storage system serial number in the subject line. The email service is available 24 hours a day, 7 days a week. If you have a service contract with an HPE-authorized service provider, contact the authorized service provider to schedule a 3PAR OS software upgrade. SP 5.0 upgrade: Upgrading from SP 4.5 to 5.0 is supported through SPMAINT only, and the upgrade must be performed by HPE service personnel. The SP running 4.5.0 does not support HPE 3PAR OS 3.3.1. The SP MUST be upgraded to SP 5.0.0 immediately following the upgrade to HPE 3PAR OS 3.3.1.

Technet24

||||||||||||||||||||

||||||||||||||||||||

You must first upgrade to SP 4.5 before upgrading to SP 5.0. Refer to the HPE 3PAR Service Processor Software 4.5 Release Notes and HPE 3PAR Service Processor Software 4.3/4.4 to 5.0 Upgrade Instructions for upgrade information and instructions. You must first verify that the HPE 3PAR OS is running version 3.3.1 before proceeding with this SP upgrade. Use SPMAINT with the command line option =1.16.6==> Upgrade SP to version 5.0 to upgrade the SP to SP-5.0.

Starting HPE 3PAR OS software update from the SP Users are notified when software updates are available, release notes link will open a new window. User can initiate the update by clicking on the Update link, as shown in Figure 8-39.

Figure 8-39 3PAR Service Console Update steps: Select the update package to install. Latest is the default. Click on the “Release notes” link to view the release notes. Brings up release notes from software portal in separate window.

||||||||||||||||||||

Load software packages in the same way as on the Update SP dialog. Selects the update method: –

Online (only option for patches)

–

Advanced Online update (only available to hpesupport and hpepartner)

–

Offline update (only option for downgrades and revert OS)

||||||||||||||||||||

Choose whether to run Admit Hardware (admithw) after the update. (Enabled by default.)

Update HPE 3PAR OS Figure 8-40 shows update details: List of actions (scripts, workarounds) executed during the update. Hidden by default except for advanced updates. For advanced updates, some actions might be optional (to be run by default).

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 8-40 Online update Click “Run Checks” to run the System Readiness Checks, as shown in Figure 8-41. Checks are required before starting update. In the case of a warning, the user is allowed to continue with the Update. In the case of a failure, the admin user is not allowed to continue. Hpesupport and hpepartner can still proceed but will be warned. The user can click on any of the checks to get more details. Details are automatically shown on any check that does not pass. Status of checks: Not checked, Passed, Failed, Failed (Temporarily), and Warning. Checks that get a temporary failure will be run

||||||||||||||||||||

||||||||||||||||||||

again after a delay specified by the 3PAR OS, up until a maximum time limit also specified by the 3PAR OS. The message box at the bottom of the dialog can be expanded if necessary.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 8-41 System Readiness screen 3PAR Service Consoles displays a screen once all phases of the updated finishes, as

||||||||||||||||||||

||||||||||||||||||||

shown in Figure 8-42.

Figure 8-42 HPE 3PAR OS update complete

Optional 3PAR OS and SP customer self-update HPE 3PAR OS 3.1.3 and 3PAR SP v. 4.2, 3PAR StoreServ 7000 and 8000 customers have the option of upgrading the 3PAR OS and SP software. Customers cannot upgrade their storage systems to 3PAR OS 3.3.1; HPE support must be contacted. SP v 5.0 is required for 3PAR OS 3.3.1 and newer. HPE policy regarding customer entitlement to 3PAR software updates and installation of the updates is that customers must maintain an active HPE Support Contract to be entitled to new software versions. HPE recommends engaging HPE or an authorized service provider to perform the upgrades if you are unfamiliar with the self-upgrade process. The Customer Self Update process is NOT supported on 3PAR StoreServ 10000 or 20000 Storage systems. If the 3PAR StoreServ Storage system is running 3PAR OS 3.1.2 or earlier, contact HPE or an HPE-authorized service provider to plan 3PAR OS and SP software updates. The Customer Self Installation option is available only for initial installs on 3PAR StoreServ 9000 arrays, not for upgrades. Customer Self Upgrade (CSU) is not available for HPE 3PAR OS software on HPE 3PAR StoreServ 9000 Storage arrays. Technet24

||||||||||||||||||||

||||||||||||||||||||

Note Customer Self Repair (CSR) information is available at this link: http://www.hpe.com/support/selfrepair It is highly recommended that you review the documentation thoroughly and understand all upgrade instructions before conducting the upgrade. While customer self-upgrade is optional at this time, if after reviewing all instructions you are not confident in upgrading, please contact your service provider for installation. HPE recommends engaging HPE or an authorized service provider to perform the upgrades if you are unfamiliar with the self-upgrade process. If automatic software download from HPE is enabled on the Service Processor, the package will be staged onto the array, and a notification is sent to the user indicating that an upgrade is available. If automatic software download from HPE is not enabled, you may download the package from the software depot at http://www.hpe.com/info/hpesoftwareupdatesupport or contact your service provider for installation. The Customer Self Update process is NOT supported on 3PAR F-Class, T-Class, or StoreServ 10000/20000 Storage systems. If the 3PAR StoreServ Storage system is running 3PAR OS 3.1.2 or earlier, contact HPE or an HPE-authorized service provider to plan 3PAR OS and SP software updates. For more information about the overall upgrade process, see the 3PAR OS upgrade Technical white paper on the HPE Storage Information Library: http://www.hpe.com/info/storage/docs The Customer must sign a Self Update Agreement; this includes a reference HPE End User License Agreement (EULA) at http://www.hpe.com/info/SWLicensing

Customer responsibilities Ensure that the host and SAN environment is supported and compliant with HPE recommendations and best practices. Resolve any problems with the SAN and host environment before initiating the software update. Ensure that all relevant documentation is reviewed before initiating the software update. The documentation includes but is not limited to:

||||||||||||||||||||

–

HPE 3PAR OS Release Notes

–

HPE 3PAR Service Processor Software Release Notes

–

HPE 3PAR Upgrade Planning Guide

||||||||||||||||||||

This documentation is located on the HPE Storage Information Library at: http://www.hpe.com/info/storage/docs

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE 3PAR StoreServ Service Processor upgrade The upgrade procedure from HPE 3PAR Service Console, shown in Figure 8-43, contains the following steps: User selects the update package to install. Latest is the default. The user can then click on the “Release notes” link to view the release notes. The user can select to load a software package onto the SP, from an ISO (uploaded through browser) or from DVD (via the SP DVD drive). The user can now start the update.

Figure 8-43 Service Processor page showing a software update available The new Service Console is the interface for the physical or virtual SP, which collects data from a managed HPE 3PAR StoreServ Storage system and sends the data to HPE Remote Support, if configured. Also allows service functions to be performed by a company admin, HPE support, or an authorized service provider. SC replaces SPOCC, the GUI for SP versions 4.4 and earlier. SC functionality is similar to SPOCC.

||||||||||||||||||||

||||||||||||||||||||

Uploading an ISO image Process of uploading the ISO to Service Console is shown in Figure 8-44. Progress is indicted by the gray bar.

Figure 8-44 Uploading an ISO image ISO file needs to be selected for uploading by either dragging and dropping the file into the gray area or using the “Choose file” button to open a browser dialog, then click “Start upload”. Once the software package is loaded on the SP, the package will show up in the dropdown list of available update package and will automatically be selected. Click “Update” to start the update.

Service Processor Update page

Technet24

||||||||||||||||||||

||||||||||||||||||||

A mini web server is deployed to keep sending updates to the user’s browser while the software update is being performed. Once the update is completed, a reboot is invoked automatically after 10 seconds. Service Processor Update process is shown in Figure 845.

Figure 8-45 Service Processor Update page

||||||||||||||||||||

||||||||||||||||||||

Activity: HPE 3PAR StoreServ 7000 Storage Remove/Replace disk video Instructions: Watch the “HPE 3PAR StoreServ 7000 Storage Remove/Replace disk” video at: https://www.youtube.com/watch?v=8gZLX2eVz2U. Notes:

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check Upgrading the service processor pre-operating system 3.3.1 uses the Service Console. True False Write a summary of the key concepts presented in this chapter.

||||||||||||||||||||

||||||||||||||||||||

Summary When upgrading storage components, it is necessary to adhere to best practices. Configurations should be optimized for cost, availability, and performance. Storage should be tuned to maintain balanced system utilization. 3PAR StoreServ operating system can be upgraded via SSMC and CLI. Service Processor can be updated using the new service console.

Technet24

||||||||||||||||||||

||||||||||||||||||||

9

HPE Backup Solutions

||||||||||||||||||||

||||||||||||||||||||

LEARNING OBJECTIVES After completing this chapter, you should be able to: Discuss the key components of the Hewlett Packard Enterprise (HPE) StoreOnce portfolio. Explain the features of Catalyst and deduplication technologies. Identify features and benefits of Recovery Manager Central (RMC). Identify feature and benefits of the StoreOnce operating system. Describe the benefits of HPE StoreEver Tape solutions.

Technet24

||||||||||||||||||||

Prelearning check Peer Copy is supported with Recovery Manager Central (RMC) 4.1 and greater. Which features are supported? (Select 2.) A. Homogenous replication between StoreVirtual Virtual Storage Appliance (VSA) and 3PAR B. Heterogeneous replication between StoreVirtual VSA and 3PAR C. Unidirectional replication D. Bidirectional replication

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

HPE StoreOnce overview A single deduplication engine to drive the movement of data across the entire organization. HPE provides a unified solution for low-cost, lights-out backup of remote offices. For environments that are too small for dedicated hardware-based backup solutions, HPE offers software-based deduplication and replication. The StoreOnce deduplication engine, shown in Figure 9-1 and Figure Figure 9-2, means that data can be efficiently replicated or copied between remote data centers and regional offices and headquarters. The goal is to deliver these capabilities using a backup application to manage the entire data protection process.

Figure 9-1 HPE StoreOnce overview

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-2 HPE StoreOnce overview

Block, file, and object compared In today’s storage industry, different types of storage will be required for different types of data; as shown in Figure 9-3, there is no One size fits all. Getting the most efficient storage type for your data is as important as getting the right storage.

Figure 9-3 Block, file, and object compared A common analogy used in many object storage discussions is the Valet Parking analogy; in this analogy, you drive your data to a parking garage and hand the keys over to a parking valet. The Object Storage Valet disassembles your car and stores the parts,

||||||||||||||||||||

||||||||||||||||||||

for data protection, across multiple parking spots and garage floors, along with encoded spare parts. That means if a parking spot, which is a server node, or a whole garage floor, which would be multiple server nodes, should fail, your original data is reassembled from the remaining data and the spare parts. In some senses, this is a little like how Redundant Array of Inexpensive Disk (RAID) works for hard disks. But unlike RAID for disks, your protection is spread further across multiple servers. And there is much less impact on performance should you have to bring a new node online, or add nodes for expansion because the rebuilding or rebalancing functions are parallelized across multiple high-performance server processors.

HPE StoreOnce portfolio The StoreOnce Backup family provides automated backup and disaster recovery (DR) operations, combining disk backup features with secure data retention and built-in data encryption for data at rest. StoreOnce Backup seamlessly integrates with Data Protector or current backup applications to provide flexible integration into storage area network (SAN), virtualized, and other environments. StoreOnce deduplication also enables network efficient off-site data replication. All StoreOnce Backup systems, shown in Figure 9-4, use StoreOnce data deduplication to significantly reduce the amount of data that needs to be replicated, enabling the use of lower bandwidth and lower-cost links to transmit data off site. StoreOnce replication helps customers achieve cost-effective centralized backup from remote sites or branch offices and delivers a consolidated DR solution for data centers.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-4 HPE StoreOnce portfolio StoreOnce Backup systems are ideal for remote offices because they provide a local backup target and an efficient deduplicated local data repository. Customers can choose to use the HPE StoreOnce VSA or dedicated appliances such as the HPE StoreOnce 3100. This choice depends on the infrastructure, performance, and management requirements of remote office deployments. For example, if a customer wants a software-defined storage solution, StoreOnce VSA is ideal. You can also build the solution using dedicated hardware appliances, providing a combination of software and hardware solutions. Comparison between various StoreOnce models is shown in Figure 9-5.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-5 Virtualized Cut storage costs by 65%, rackspace by 50%, and power by 70% with a softwarebased, hardware-agnostic VSA that eliminates the need for dedicated hardware or backup appliances. Consolidated Save operational and management time by managing remote backup and replication from a single, centralized location using the backup software of your choice.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Federated Reduce risk and increase agility using HPE’s unique ability to conform to whatever backup paradigm you need and then securely replicating data without ever having to rehydrate. Standardize on a single deduplication solution across the enterprise for cost-optimized data replication. Multitenant Enables small and midsize service providers to deliver managed services that meet customer data protection needs while reducing CAPEX (capital expenditure) and OPEX (operating expense). Cost-effective backup as a service offering without the need for dedicated hardware. More scalable options: A new 32TB StoreOnce VSA has been introduced, for users who want more than 20TB but who do not yet require as much as 50TB. StoreOnce VSA provides you with the flexibility to deploy as per your data protection requirements, with licensing options available at 4TB, 10TB, 20TB, 32TB, and 50TB capacity points. To enable scalability, a 4TB license can be upgraded to 10TB, 10TB to 20TB, 20TB to 32TB, and 32TB to 50TB. What is new

Mixed couplet support within the HPE StoreOnce 6600, customers can now add a 6600 couplet to an existing StoreOnce 6500 appliance. HPE StoreOnce 5500 and 6600, now with 3-year warranty. Catalyst Copy over Fibre Channel (FC), use FC as a protocol to transmit copies in addition to backup. Deduplicate anywhere via physical or virtual appliances—at the application source, the backup server, the target appliance, or directly from 3PAR StoreServ via StoreOnce Recovery Manager Central (RMC). Deduplicate where it makes sense for your business, not where technology vendors’ limitations mandate, saving time and money. Federated Deduplication is available across the HPE StoreOnce Systems portfolio— including dedicated appliances, virtual backup solutions, and on media servers and application servers with HPE Data Protector, Veritas NetBackup, and Backup Exec software via OST (Open Storage Application Programming Interface—API), Veeam, and BridgeHead software. HPE StoreOnce also supports critical business applications via no cost Catalyst plug-

||||||||||||||||||||

||||||||||||||||||||

ins for Oracle RMAN, SAP HANA, MS SQL, and SAP on Oracle. HPE StoreOnce Federated Catalyst in a multinode system allows Catalyst stores to span nodes simplifying backup management and optimizing available storage in large environments.

StoreOnce Deployment planning and sizing considerations Important considerations when planning and sizing StoreOnce solution for a customer include: What operating systems are running on Media Server/Media Agent? Is the environment FC or Ethernet? What speed is their SAN/local area network (LAN)/wide area network (WAN)? Does the environment have any known bottlenecks? What data types does the customer plan to backup to StoreOnce? Do they have any encrypted/compressed data? Construct a topology of the customer’s environment What problem is the customer trying to solve? What is the customers’ priority? Dedupe Ratio? Performance? Ease of management? Capacity? All the above? Always consult the Data Agile Support matrix, If it is not listed it is an unsupported configuration. Missing this step could lead to an unusable StoreOnce deployment. It is better to identify early whether the configuration is supported or not.

Choosing the most appropriate StoreOnce system Figure 9-6 shows key features and use cases for network-attached storage (NAS), virtual tape library (VTL), and StoreOnce Catalyst.

Technet24

||||||||||||||||||||

Figure 9-6 Comparing StoreOnce backup interfaces Installation planning—Data center considerations: Does the customer have the required space on the data center floor for StoreOnce rack(s)? Is the customer likely to expand their StoreOnce in the future? If so, leave floor/rack space for expanding StoreOnce. Does the data center meet StoreOnce power requirements? Check that correct power distribution units (PDUs) have been ordered. Installation planning—What StoreOnce network configuration will the customer use? Will management and backup Ethernet traffic be combined? Will management and backup Ethernet traffic be separated? Will a 1 Gb or 10 Gb Ethernet network be used? –

||||||||||||||||||||

Recommendation: 1GbE for management, 10GbE for data

Will vLAN tagging be used? Will port bonding options of Mode 1 (active/passive), Mode 4 (Link Aggregation Control Protocol—LACP), or Mode 6 (active/active) be used? Port bonding options of Mode 1 (active/passive), Mode 4 (LACP), and Mode 6 (active/active) are available? Licenses: Consider security

||||||||||||||||||||

||||||||||||||||||||

Storage Capacity Licensing, Catalyst/Replication (simplified single license from 3.9 onward). Ninety-day, instant on demo license is available via command line interface (CLI) (#license add demo). Simple Network Management Protocol (SNMP)/Email Alerting and Insight Remote Support (IRS) if required. Setting up alerting is often a missed stage, but it is vital for customers to be able to identify any issues early. Remote Support Service Tools and Technical Service (STaTS). Remote Support (STaTS) generated cases have correct SKU (stock keeping units)/SN values. Optional hardware configuration (Flex IO): Make sure to add optional hardware to the order. Make sure to select the appropriate host bus adapter’s (HBA) fulfilling environmental and throughput requirements. IP address allocation. If possible, for multinode, allocate IP addresses upfront, during an initial cluster installation. Additionally added couplets will then automatically make use of the preallocated IP addresses without needing to disrupt the current configuration.

Configuration best practices—General Choosing the number of backup devices (stores): Performance is attained by sharing the work load across multiple backup devices (VTL/NAS/Catalyst Store). The number will depend on the customer deployment, but recommendation would be between two and eight backup devices per appliance/node. If performance is crucial, single-store configurations might hit performance limitation. Configure a large block size min 256KB in backup application. Separate backup devices for different data types—to maximize dedupe ratios: Common data types dedupe better against each other more than foreign data types.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Performance is attained by improving dedupe ratios. Separating windows of activity—Backup/Replication/Housekeeping/tape offload: Each operation puts a heavy work load on the StoreOnce. Separating the activities into different windows will improve performance. Make use of StoreOnce blackout windows for Housekeeping, Replication, and Catalyst copy. Confirm you are running the latest StoreOnce software code: Go to http://www.hpe.com/support. Click at Product Support “HPE Servers, Storage and Networking”. Enter “HPE StoreOnce Systems”. Click at “Driver, Software & Firmware”. Choose StoreOnce model out of the list. Choose operating system “OS independent”. Expand “Software” section. Choose StoreOnce Software version to be downloaded, click “Obtain software” and follow instructions.

Performance: Understand performance curve progression Figure 9-7 shows StoreOnce performance curve progression of high and low bandwidth: High-bandwidth backup (HBW): VTL, CIFS (Common Internet File System), NFS, Catalyst target side dedup Low-bandwidth backup (LBW): Catalyst source/server side dedup

||||||||||||||||||||

||||||||||||||||||||

Figure 9-7 Performance: Understand performance curve progression Curve progression shown in dependency of StoreOnce capacity expansion kits. Tested with a total of 48 concurrent streams (5100, 5500, 6600 node) Tested with 10% data change rate In our example, high-bandwidth backup suggests that there is no significant increase in performance when adding additional capacity expansion units. However, the low-bandwidth backup clearly shows a steady increase in performance, in this example, when more capacity expansion units were added.

Performance considerations Performance considerations, as shown in Figure 9-8, include: It is recommended to use a minimum of three streams, though six is better. Distribute concurrent running streams evenly to all available stores. Run multiple backups in parallel to improve aggregate throughput to StoreOnce. Use multiple concurrent streams per backup job. Benefit from using minimum two to three Stores (VTL, NAS, Catalyst) vs single large Store.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-8 Performance considerations

Separating windows of activity To improve performance of StoreOnce solution, separate different types of activities: backup, replication, housekeeping, and tape offload. The scenario depicted in Figure 9-10 shows a bad tasks planning, which you should try to avoid.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-9 Overlapping jobs of different types The scenario depicted in Figure 9-11 shows a good tasks planning. Configure blackout windows to get a similar result.

Figure 9-10 Different job types running in separate time ranges Housekeeping runs as a background task, and it can still interfere with backup and replication performance. Past and current housekeeping state:

Technet24

||||||||||||||||||||

Before v2.1 Housekeeping would only run immediately: –

When a virtual tape was overwritten and unloaded

–

When a NAS file was closed

||||||||||||||||||||

Now, housekeeping can be monitored and configured to run at specific times. You must allow time for housekeeping each day, so StoreOnce can clean up after backup/overwrite operations.

Do not send multiplexed data to StoreOnce backup systems

Figure 9-11 Multiplexed vs. concurrent streams Multiplexed data does not deduplicate that well, and there is no reason to use it with StoreOnce. Figure 9-11 shows multiplexed and concurrent streams. In Data Protector, set concurrency to “1” to avoid multiplexing, as shown in Figure 912. This setting needs to be done for each single tape drive in Data Protector under tape drive properties.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-12 Setting the concurrency value to 1

Network interfaces considerations Choose the appropriate interface 1 or 10GbE for required throughput. It is recommended to use 1GbE for management and 10GbE for data. Choose the appropriate network bonding mode (1, 4, or 6) that fulfills requirements. Up to 128 vLAN can be configured per StoreOnce model.

Additional considerations Backup performance depends on disk performance, backup client, infrastructure bandwidth, throughput performance of the backup server, and backup target. The slowest device or the medium with the smallest bandwidth dictates backup performance. Sequential backup is shown in Figure 9-13.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-13 Sequential backup example Multiple disk agents, as shown in Figure 9-14, can improve performance in many cases. In this case, each disk drive creates a minimum of one backup stream that is able to run concurrently. In the StoreOnce backup system, you are creating VTLs with multiple virtual tape drives at no extra cost compared to physical tape library.

Figure 9-14 Parallel backup example As the StoreOnce backup system requires multiple concurrent backup streams for best performance, it is advised to go for a setup that uses concurrent running backup streams.

||||||||||||||||||||

||||||||||||||||||||

Choosing the number of backup devices—StoreOnce single node VTL/NAS/Catalyst recommendations: Send the same data types to the same device configured on the StoreOnce system, to maximize deduplication ratios. Run multiple backups in parallel to improve aggregate throughput to a StoreOnce Appliance, as shown in Figure 9-15. To be able to reach an acceptable performance, work with a minimum of two stores and split the number of reference streams per appliance equal to each store.

Figure 9-15 StoreOnce single-node example

Choosing the number of backup devices—StoreOnce single-node (6000 series) VTL/NAS/Catalyst recommendations: Use data segmentation strategy. For higher throughput, create multiple stores (VTL/NAS/Catalyst) for the same data type on different nodes, as shown in Figure 9-16. Run multiple backups in parallel to improve aggregate throughput to a StoreOnce Technet24

||||||||||||||||||||

||||||||||||||||||||

multinode and to each single node. To be able to reach an acceptable performance, work with a minimum of two stores per node and split the number of streams per node equally to each store.

Figure 9-16 StoreOnce single-node example

Choosing the number of backup devices—StoreOnce multinode Federated Catalyst recommendations: Consolidate dedupe stores and simplify management with automatic capacity balancing across nodes, as shown in Figure 9-17. It spans across up to eight nodes. To be able to reach an acceptable performance, work with a minimum of two Federated Catalyst Stores per system and split the number of streams per node equally to each store. A single Federated Catalyst Store can handle a maximum of 512 streams.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-17 Multinode with Federated Catalyst example

StoreOnce VSA HPE StoreOnce VSA delivers the cost efficiency of deduplicating backup storage in a software-defined form factor for VMware, Hyper-V, and Microsoft Azure environments. The space saving, high backup performance and deduplication-optimized replication of HPE StoreOnce are delivered in a virtual appliance form factor. StoreOnce VSA features: Delivers 1 to 50TB of raw storage—cost/flexibility/security. 1TB free demo license. No support for VTL on FC. Fully integrated with HPE StoreOnce Enterprise Manager (SEM) for management and reporting. Supported on VMware and Microsoft Hyper-V platforms. Provides up to 16 integrated Small Computer System Interface (iSCSI) VTL targets,

Technet24

||||||||||||||||||||

NAS shares, or StoreOnce Catalyst Stores. Fully compatible with the StoreOnce hardware-based appliances. Is able to replicate VT/NAS or use StoreOnce Catalyst copy. StoreOnce VSA can act as a replication target for ONE StoreOnce appliance or StoreOnce VSA source. Additional specifications are shown in Figure 9-18.

Figure 9-18 StoreOnce VSA specifications Requirements for the StoreOnce VSA: Virtual CPU: Minimum 2x vCPU –

Above 4TB storage capacity 4x vCPU recommended

–

Above 10TB storage capacity 4x vCPU recommended

Virtual RAM: Minimum 16 GB of vRAM –

Above 4TB storage capacity 24 GB of vRAM recommended

–

Above 10TB storage capacity 32 GB of vRAM recommended

Virtual disk: 50 GB thin provisioned system disk for ESXi and 50 GB fixed system disk for Hyper-V –

||||||||||||||||||||

Data disk of any size in multiples of 1TB, thick provisioned lazy zerced of thin provisioned for ESXi and fixed or dynamically expanding for Hyper-V

Virtual NIC: 2x 1GbE or 10GbE virtual network interfaces Sufficient SCSI port available on the controller to support the number of virtual disks that will be deployed

||||||||||||||||||||

||||||||||||||||||||

StoreOnce VSA: Licensing information The StoreOnce VSA is fully functional from its first installation with a 60-day instant-on period. If no license key is added within 60 days of start up, all backup targets become read-only. Once a license key is added, full functionality is returned. If a license to use (LTU) reaches the end of its term, all backup targets become read-only. Once a valid license key is added, full functionality is returned. The StoreOnce VSA license enables use as a replication target and does not require an additional Replication license. If you intend to replicate to VTL and/or NAS targets on a StoreOnce appliance, a Replication license will need to be installed on the target appliance. The StoreOnce VSA license enables creation of Catalyst Stores and execution of Catalyst Copy operations and does not require an additional Catalyst license. If you intend to use Catalyst Copy between a StoreOnce VSA and a StoreOnce appliance, the StoreOnce appliance will need a Catalyst license installed. The StoreOnce VSA license enables on data at rest encryption, data in flight encryption, and secure erase and does not require an additional Security Pack license. The StoreOnce VSA leverages the flexibility of virtual infrastructures by packaging the StoreOnce deduplication technology in virtual appliances for VMware, Microsoft Hyper-V, and Ubuntu KVM environments. StoreOnce VSA is licensed at 4TB, 10TB, 20TB, 32TB, and 50TB capacity points and can be deployed on any industry-standard server with the appropriate specification. A 1TB StoreOnce VSA can also be used free of charge.

StoreOnce VSA: SKU information Licenses are purchased with a three-year, five-year fixed or perpetual term. For termbased (three or five years) licenses, the fixed term can be extended by the addition of another term-based license or a perpetual license. A license to extend an existing fixed term must match the capacity of the existing base license. If you attempt to activate a license for an already-licensed StoreOnce VSA that has a nonmatching capacity license, the activation will fail. The fixed term starts from the day the license is redeemed from the HPE licensing portal not the day the license is added to the StoreOnce VSA. Further Considerations for Perpetual Licenses: Perpetual licenses are only supported with StoreOnce VSAs running 3.13.2 or newer software. Foundation support is included for three years. Support services are available to Technet24

||||||||||||||||||||

||||||||||||||||||||

purchase to extend this support to four or five years and to upgrade the support coverage and response level.

Backup target emulation types—VTL VTL, shown in Figure 9-19, includes following features: Seamless integration into most backup application. Emulated tape libraries: MSL G3 (2x24, 4x48, 8x96), EML-e, ESL-e, D2DBS Generic, IBM-TS3500, IBM-TS3500 IBMi. Emulated tape drives: HPE LTO-2 (Linear Tape Open), -3, -4, -5, -6, Ultrium VT (virtual generic Ultrium device), IBM LTO-3, IBM LTO-5 IBMi. Flexible cartridge sizes, slots, and emulation types. Library changer device can be mapped out via multiple FC port for redundancy. Up to 500 tape drive per library (5500/6600). Non-Deduplication mode for pre-encrypted or pre-compressed data (also legal compliance in some regions). Enable encryption for data at rest. Secure Erase functionality (up to seven overwrite passes).

Figure 9-19 Backup target emulation types—VTL

||||||||||||||||||||

Backup target emulation types—NAS Considerations for using StoreOnce NAS target, as shown in Figure 9-20: CIFS—Server name: Should match hostname (small change to default hostname made to ensure this). Authentication method: None, User, AD Workgroup/Domain name information. User List: List of usernames which can be given access to shares. NFS—Server name: Same as for CIFS. Host List: List of hosts which can be allowed access to shares. Two versions of NAS share available, Type 1 and Type 2. “Type 1” store has the following: –

Maximum file limit is 25000.

–

The first 24 MB is left undeduplicated (intended for independent software vendor [ISV] R/W).

“Type 2” store has the following: –

Maximum file limit is 1 million

–

The first 1 MB is left undedulicated (intended for ISV R/W).

–

Optimized to improve performance with CommVault Simpana backup software.

||||||||||||||||||||

Figure 9-20 Backup target emulation types—NAS Security feature Data at Rest encryption for NAS targets on all StoreOnce products

Technet24

||||||||||||||||||||

requires Secure Pack license. CommVault Simpana aware NAS shares are optimized for deduplication ratio and performance. The StoreOnce deduplication engine was designed for a sequential workload, with NAS there is the possibility of overwriting data before the end of a file—this is termed Write-in-place (WIP). To allow this happen, two steps were taken: Since most WIP is done at the start of a file (for example, ISV header catalogues), the first 1 or 24MB of each file is not deduplicated and therefore “random access”. Any WIP that occurs in the deduplicated portion of a file is dealt with by putting the new data in a separate file and referencing it so that it is retrieved during restore. Deduplication of a NAS share can be completely disabled at store creation time. In case you know that lots of random IO will happen, or using pre-compressed or encrypted data to be stored. User selectable non-deduping NAS shares are also useful when source side deduplication is used in backup software.

Learning check Object storage is best suited for: A. Storing and sharing files that might be accessed on a frequent basis B. Databases and other business application C. Large amounts of infrequently accessed and long-term unstructured data

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

StoreOnce Catalyst StoreOnce Catalyst software, shown in Figure 9-21, provides a simple, consistent, highperforming architecture that spans the organization. It delivers high backup speeds to meet shrinking backup windows. It also provides federated deduplication across the enterprise, including the consolidation of backup and DR from multiple remote offices to the data center. Multiple StoreOnce appliances and virtual machines can replicate to the central StoreOnce appliance. Protect data from unauthorized access through data-atrest encryption and secure erase functionality for disks that are lost, stolen, or discarded.

Figure 9-21 StoreOnce Catalyst Benefits for ISV software: This is a device that ISVs can use with none of restrictions of tape emulation and NAS shares. No unnecessary “geometry” limitations (library slots, cartridge sizes, etc.). Allows ISV software to be aware of the StoreOnce Backup system and its capabilities. Allows the backup server to support node failover in multinode systems without complex restart scripts (some ISV still have scripts) and restart from checkpoints. Use bandwidth efficient methods to move (copy) data without rehydration. This enables backup over WAN or LAN, with increased performance.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Enables deduplication in different locations (federated deduplication). Enables ISV applications control copies of backups. Copy to multiple locations and to set different expiry dates. ISV is fully aware of all copies and jobs. Catalyst provides tight integration with HPE Data Protector, fully supports Symantec OST or Veeam.

HPE StoreOnce Catalyst ISV integration

Figure 9-22 Target and source-side deduplication Without Catalyst, there are only target-based deduplication for VTL/NAS. This dedupe architect is also being called high-bandwidth backup because during a full backup, the complete data capacity will be traveling over a LAN/SAN. Also, deduplication uses a unidirectional communication, so server has to use its own CPU/memory to do data replications and synthetic full backups. With Catalyst, low-bandwidth backup, server-side/source-side deduplication are possible, so even in a full backup, only the unique blocks are sent by servers to StoreOnce. As it is a bidirectional communication, Backup Server can now offload jobs like data replication and generating a virtual synthetic full backup job to StoreOnce, and StoreOnce will inform Backup ISV after these jobs are finished.

||||||||||||||||||||

||||||||||||||||||||

Catalyst stores allow backup applications to use low-bandwidth deduplication, serverside deduplication on, or high-bandwidth deduplication, server-side deduplication off. You can manage backup and DR operations from a single console with StoreOnce Catalyst. Catalyst over FC provides all the ISV control and source-side deduplication benefits of StoreOnce Catalyst but via an FC fabric. Automated, efficient backup and DR operations with StoreOnce Catalyst include: DR plans that were not previously feasible because of lack of multisite capabilities, cost of bandwidth, and time. One-to-many DR moves data simultaneously from one site to many sites. Cascaded DR moves data sequentially from one site to any number of other sites. Alerting and trending in SEM for backup capacity management.

Multihop When the Catalyst emulation reaches the data center, the StoreOnce system does not need to rehydrate the data. However, Catalyst, which resides there, moves the duplicated dataset directly onto the DR site. This is called cascaded data movement or multihop. During this process, the backup application in the remote office tracks and controls the movement of this dataset, as shown in Figure 9-23.

Figure 9-23 Multihop

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE StoreOnce deduplication StoreOnce deduplication software simplifies the deployment of deduplication technology across IT infrastructures. With explosive data growth driving IT sprawl, deduplication technology is quickly becoming a requirement for many customers who need to reduce storage capacity. The actual data deduplication ratio depends on a number of factors including: Type of data Backup methodology used Length of time data will be stored

HPE deduplication write pipeline The goal of deduplication is to allow customers to store more data on disk for longer periods by eliminating redundant data. In general, deduplication methods fall into two categories: Post processing where deduplication is performed after backups complete. This is normally a byte comparison type process. In-line hash-based chunking where deduplication is performed as data is received from the host system. The StoreOnce deduplication technology uses asynchronous pipelined transformations to achieve the performance required, as shown in Figure 9-24. Each transformation is separated by a FIFO (first in first out buffer) and can run in parallel taking advantage of multicore CPU technology. Each VTL or NAS share has its own dedupe store held on disk. For each IO stream (to a file share or virtual tape drive) will have active threads that can run on separate CPU cores.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-24 HPE deduplication write pipeline The write pipeline processes are described as follows: Chunking: This process cuts the data stream into chunks of on average 4KB using the Two Thresholds Two Divisors (TTTD) algorithm. Hashing: Uses SHA-1 hash algorithm on each chunk. Matching: Decides whether the StoreOnce has stored the chunk before by comparing hash values. Compression: Each chunk is compressed using the LZO (Lempel–Ziv–Oberhumer) algorithm before writing to disk. Duplicate chunks are not written to disk. Sort & Save: Decides where to put the data on the disk. Note LZO is a lossless data compression algorithm that is focused on decompression speed. (March 2018, Wikipedia.org)

Technet24

||||||||||||||||||||

||||||||||||||||||||

Note TTTD algorithm must specify the minimum and maximum sizes of chunks (two limits). From the first byte of the file, the algorithm reads byte by byte until it reaches the minimum size of the chunk. Then, it computes the hash of the chunk and calculates the module of the hash with a main and a secondary value (two dividers). If the result is equal to the first divisor, it means that the chunk must begin there. (March 2018, http://stacksync.org/desktop-client/chunking) HPE StoreOnce deduplication is performed as the data is received from the host system and uses in-line hash-based chunking technology. Deduplication appliances have to provide target devices for the host systems, and these typically are Virtual Tape or NAS. HPE has an additional “target” known as StoreOnce Catalyst. StoreOnce deduplication software simplifies the deployment of deduplication technology across IT infrastructures. With explosive data growth driving IT sprawl, deduplication technology is quickly becoming a requirement for many customers who need to reduce storage capacity.

Federated deduplication HPE federated deduplication, shown in Figure 9-25, uses a common StoreOnce deduplication algorithm that allows users to deploy solutions across the storage infrastructure. It provides deployment independence for deduplication because the common deduplication engine enables the native communication and movement of data across various HPE systems without deduplication. This increases efficiency, especially because data moves between sites over a low-bandwidth connection. It also provides flexibility in deduplication strategy. Federated deduplication allows data reduction to occur in Data Protector software or in the StoreOnce family of appliances.

Figure 9-25 Federated deduplication

Deduplication optimization HPE created a highly optimized deduplication approach that introduces time- and

||||||||||||||||||||

||||||||||||||||||||

space-saving techniques. The goal is to eliminate as much redundancy as possible in data inspection, maintain a small index, and deliver the fastest performance. The deduplication approach from HPE focuses on two components: an average variable chunk size of 4K and a sparse index. HPE adaptive micro-chunking uses variable-length data segments or chunks. The backup stream is broken down into approximately 4K variable-length segments that are examined for redundancy against previously stored information. Smaller segments mean that there are more chunks and index comparisons, increased potential to locate and eliminate redundancy, and higher reduction ratios. Comparative solutions use block sizes that range from 8K to 32K. The trade-off with small chunk sizes is a greater number of index lookups, which could mean slower deduplication performance. However, HPE Labs developed predictive acceleration technology to maintain performance and reduce RAM requirements. By using a subset of key values stored in memory, StoreOnce determines a small number of sequences already stored on disk that are similar to any given input sequence. HPE refers to this as sparse indexing. Each input sequence is only duplicated against those few sequences. This minimizes disk IO and uses less disk and little memory, creating more efficiency and enabling faster ingestion and restoration of data. The HPE approach accelerates reads and writes, delivering rapid ingestion rates of up to 28TB per hour. Predictive acceleration has allowed HPE to reduce demands on RAM by up to 50% compared to competitive solutions.

Federated StoreOnce Catalyst HPE StoreOnce Backup with HPE StoreOnce Catalyst supports a federated deduplication solution, illustrated in Figure 9-26, providing disk-based backup for IT environments, from small remote sites to large enterprises. StoreOnce Catalyst helps customers: The architecture is an extension of the HPE StoreOnce Catalyst implementation and protocol. Designed of HPE multinode appliances (B6200, 6500, and 6600). Catalyst “client” software is embedded in HPE DataProtector code or in a “plug-in” module for Symantec NetBackup. Each StoreOnce “service set” that runs on a “node” hosts one or more federated stores. The catalyst “client” running on a backup server presents a single logical federated catalyst store. Catalyst client aggregates data from multiple nodes.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Supported on DataProtector and Symantec NetBackup, RMAN and SQL plug-ins also supported.

Figure 9-26 Federated StoreOnce Catalyst

StoreOnce Catalyst principles When the Catalyst emulation reaches the data center, the StoreOnce system does not need to rehydrate the data, Catalyst moves the duplicated dataset directly onto the DR site. Restoring data from the DR site to the data center is faster than with other emulations because it is the same duplicated dataset that you started with. Figure 9-27 shows an architecture of StoreOnce Catalyst.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-27 StoreOnce Catalyst principles

STaTS support for customers with 3PAR and StoreOnce STaTS, shown in Figure 9-28, is one of the methods which StoreOnce Appliances (and RMC) can “phone” home to HPE. Incidents are evaluated and passed then to level 1 case management who can take the necessary action. STaTs also gathers statistical information on performance. HPE customers can register and view their own systems through the StoreFront application. HPE employees have access to StoreVista that is capable of running more comprehensive reports.

Figure 9-28 STaTS support for customers with 3PAR and StoreOnce

Technet24

||||||||||||||||||||

||||||||||||||||||||

STaTS allows authorized HPE personnel to access historical performance, event, and configuration data from selected HPE Storage products. STaTS data is guarded and always treated as HPE Confidential data. The data retention can allow for quick analysis of a customer’s storage, access to event data from previous data captures. Data stored within the STaTS database can only be accessed by securely authenticated and authorized HPE employees associated with the users’ account.

StoreVista individual system detail and performance history Every call-home–enabled StoreOnce device can be reviewed online via StoreVista. Including current config and historical capacity/performance details, as shown in Figure 9-29.

Figure 9-29 StoreVista individual system detail and performance history StoreOnce Remote Support will automatically be enabled. Customers who do not want remote support will need to manually turn it off.

||||||||||||||||||||

Nimble SF-Series Secondary Flash Array Nimble SF arrays (SFA) offer number of features that make them perfect candidate for backup storage. Features of Nimble SF-Series, shown in Figure 9-30, allow customers to: Put backup data to work –

Secondary storage that does real work—Zero-copy cloning and flash performance let you use your backup data for development/test, QA, analytics, and more.

–

Data reduction without penalty—Always-on, inline deduplication, and compression increases effective capacity without impacting backup and recovery.

–

Peace of mind—Flash performance to verify backups quickly and easily, plus 99.9999% measured availability for when you need to run production workloads too.

Reduce backup and recovery windows –

Near-instant restores—Access files, VMs, applications, and entire systems directly on the SFA and/or rapidly copy them back to primary storage.

–

Near-instant disaster recovery—Fail over to the SFA to resume running production workloads at speed.

–

Near-instant, snapshot-based backups—No more backup windows and zero impact to hosts.

Simplify operations –

Easy availability for virtualized environments—Built-in integration and validation with Veeam Availability Suite delivers broad compatibility, whether you are using primary storage arrays from HPE or from another vendor.

–

Worry-free operations—InfoSight Predictive Analytics anticipate and prevent issues, automatically predicting and resolving 86% of problems before you even know an issue exists.

–

Simple acquisition and refresh—All Nimble Storage arrays feature the Timeless Storage guarantee, which means that hidden costs and forklift upgrades can become a thing of the past.

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-30 Nimble SF-Series advantages

Veeam integration Veeam understands that traditional backup methodologies are fundamentally broken. We have only 24 hours a day to perform the backup and our data continues to grow, scraping data from a primary subsystem impacts the performance of that array and managing the process to meet an insurance policy is not the best use of our time and resources. Deduplication backup to disk targets is only a bandaid to the larger problem, the problem of the backup window. As part of Veeam 9.5, Veeam and Nimble have created integration, shown in Figure 931, that enables Veeam to catalogue Nimble Storage snapshots, understand the contents of those snapshots and is replica aware. In other words, Veeam will manage your data protection but leverage Nimble SmartSnap and SmartReplicate technologies. Veeam gives you the ability to perform file level recovery from any site that it protects and totally eliminates the need now for backup to disk. All of your recoveries, whether it be a full volume, array, or file, will always be retrieved from Nimble Storage (near instantaneous), but you will not longer have the flow of data to and from a backup target. If long-term retention (say beyond six to nine months) is required, you can still use Veeam to do that full backup every month or quarter or whether you determine is appropriate to meet your data protection requirements.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-31 Veeam integration

Technet24

||||||||||||||||||||

Learning check Low-bandwidth deduplication is: A. Server side B. Target side

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Activity: StoreOnce Systems to deal with data growth video Instructions: Navigate to the following website: https://www.hpe.com/uk/en/storage/storeonce.html#Resources. Scroll down the page to the “StoreOnce Systems to deal with data growth” video, and click Watch the video, as shown in Figure 9-32.

Figure 9-32 Watch the Video Notes:

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Recover Manager Central Features of RMC include: Express Restore Guided workflow Protection policy for RMC core and RMC-S Peer Copy enhancements Catalyst Copy for RMC-SAP/HANA via UI Suspension of schedules Deployment of RMC VM to NFS datastores

RMC components HPE StoreOnce RMC serves as the single integration and control point for data protection. Leveraging primary 3PAR StoreServ Snapshot technology and StoreOnce Catalyst Backup for converged data protection that combines the simplicity and performance of snapshots with the reliability and storage efficiency of deduplicated backup. Recovery Manager Central solution is built on different hardware components and features which work together, as shown in Figure 9-33. To provide an effective backup solution without the use of an ISV providing the backup software. The end result of this collaboration would be that the snapshots that are created on the HPE StoreServ are then backed up to the Catalyst Store on the HPE StoreOnce.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-33 RMC components To get the idea about what Catalyst Store represents, you have to know that each backup job requires the destination. Traditionally, destination could be the tape device, VTL, or shared folder. Backup engine capable of using the Catalyst Store as the backup destination benefits from its deduplication function known as Federated Deduplication, which might be deployed across a storage infrastructure. The use of a common deduplication engine enables the native communication and movement of data across the various systems without rehydrating the data. Therefore, the subsequent snapshots moved to the Catalyst Store would be automatically deduplicated. Virtual volumes are created on the HPE StoreServ, and they could belong to the Virtual Machines or to the individual servers within the data center environment. Snapshots of these Virtual volumes are created on the 3PAR using the 3PAR snapshot technology. The VMware ESXi host is used to support customer’s virtual machines and special virtual appliance executing the Recovery Manager Central engine. Currently, HPE supports the ESXi versions 5.5, 6.0, and 6.5. The Recovery Management Central GUI could be accessed over the network and is used for monitoring and initiating the Express Backup procedure. Considerations include: RMC is a CentOS-based Virtual Appliance (VA). Deploy RMC either on a Hyper-V VM or VMware VM. No root login access given to the VM. Configuration is done through the RMC configuration GUI.

||||||||||||||||||||

||||||||||||||||||||

Additional RMC services need to be configured to use the required devices/servers.

RMC licensing for 3PAR 8000/9000/20000 All requisite RMC licenses are included by default as part of the 3PAR all-inclusive single-system software license suite, for the 3PAR 8000, 9000, and 20000 series of arrays. (The 3PAR Virtual Copy software license is a prerequisite for RMC to work, but the Virtual Copy license is also part of the all-inclusive suite.) In other words, the RMC-based snapshot management functionality is available for no additional cost to all 3PAR 8000, 9000, and 20000 customers, without any additional licensing requirements. If RMC, shown in Figure 9-34, is used to manage snapshots of remote copy volume groups across two 3PAR StoreServ arrays under a remote copy replication relationship, then a 3PAR Remote Copy software license is a prerequisite, which is part of the 3PAR all- inclusive multisystem software license suite.

Figure 9-34 RMC licensing for 3PAR 8000/9000/20000 Peer Copy, however, does require the 3PAR Remote Copy software license and the StoreVirtual Remote Copy license to be enabled as a prerequisite on the corresponding array platforms. To unlock the Express Protect feature to carry out direct backups from supported storage

Technet24

||||||||||||||||||||

||||||||||||||||||||

arrays to HPE StoreOnce Backup systems using RMC, the only license required is an HPE StoreOnce Catalyst license. The 3PAR Remote Copy license is available as part of the 3PAR all-inclusive multisystem software license suite on the 3PAR 8000, 9000, and 20000 series. The StoreVirtual Remote Copy license is also included with the product as part of the allinclusive license. The 3PAR all-inclusive suites are not available for the legacy 3PAR 7000 and 10000 series of arrays. For customers using these older array models, the RMC snapshot management functionality can be licensed by purchasing a corresponding RMC Suite SKU (Please see previous section for SKU details). These Suite SKUs enable all application plug-ins including RMC-V, RMC-S, RMC-O, as well as RMC-SH and are licensed per array frame regardless of amount of data protected. Customers ordering HPE Hyper Converged 250 System for VMware vSphere on or after 30 August 2016 are already licensed for RMC. All requisite licenses to enable the RMC-V functionality to create, schedule, manage crash-consistent or application-consistent snashots on the StoreVirtual data services running on the HC 250 System for VMware vSphere is included in the HC250 product bundle itself. No separate license is required.

Monitoring RMC Appliance—Dashboard The RMC Dashboard, shown in Figure 9-35, provides a graphical representation of the general health status of several managed resources in your data center. From the Dashboard, you can immediately see the areas that need your attention. For direct access to resources needing your attention, select the resource name for a filtered view. To view the dashboard graphs, select Dashboard from the main menu. Only those resources you are authorized to view or manage appear on the Dashboard.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-35 Monitoring RMC Appliance—Dashboard The Dashboard screen details provide resource-specific donut graphs on total number of resources, the status of the resources, and the number resources in each state. The Summary option displays a summary view of data center capacity and health information. The information displayed is spherical and graphical. Click the spherical shape to navigate to HPE Storage System Page. The details option displays the bar graph, the dedupe ratio for each backup device. When a backup device is selected from the list box, the graph is updated to show the dedupe ratio of each store for the selected backup device. The Summary option displays the number of currently running backup jobs and the number of failed jobs in in the last 24 hours. The colors of the donut graph represent the status of the resources: Green—A healthy status. Red—A critical condition that requires your immediate attention. Gray—The resources are in an unknown state. Black—Resource instances reporting status other than OK, Warning, or Critical. For example, Disabled or Unknown.

Peer Copy Heterogeneous Replication between StoreVirtual VSA and 3PAR, as shown in Figure 936.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-36 Peer Copy What is it? Volume-level data copy enabled from 3PAR to SV VSA and from SV VSA to 3PAR Suggested Use Cases Replication of VSA in remote sites to 3PAR in central data center for business continuity Replication of 3PAR StoreServ 8200 to VSA for low-cost 3PAR DR The new data copy feature within RMC allows customers to bidirectionally replicate data volumes between StoreVirtual VSA and 3PAR arrays—eliminating the need to use host-specific or hypervisor-specific tools or expensive network-based appliances. RMC orchestrates and moves data from a source to destination storage system. It automates the data copy process and makes the volume available for use after the copy process without any need to convert the data. Enhancements in RMC4.1 Preseeded Remote Office/Branch Office (ROBO)—The ability to import a preseeded volume to create a replication set. User editable replication stream performance enhancements. Data mover log cleanup. Handling of WSAPI being down. Support for SV VSA 12.7. Enhancement of Schedule UI workflow in Replication. Enhance cancel replication, cancellation should be allowed in all prereplication stages. Consolidated view of replication set. Allow multiple delete of replication sets. Replicate SV VSA from remote sites to central 3PAR for business continuity.

||||||||||||||||||||

||||||||||||||||||||

Replicate 3PAR to SV VSA for low-cost 3PAR DR or use it for migration scenarios. Heterogeneous Replication between StoreVirtual VSA and 3PAR. Peer Copy in RMC 4.0 is not supported using StoreVirtual VSA 12.7. StoreVirtual VSA 12.7 is supported for Peer Copy when using RMC 4.1. Not Supported on StoreVirtual/LeftHand hardware appliances. Not Supported on HC 380 appliance. Needs 3PAR Remote Copy License.

Express Protect Express Protect is a fast backup, shown in Figure 9-37. Only Snap differential data is copied, and then a Synthetic Full backup image is created on StoreOnce.

Figure 9-37 Express Protect

Requirements for Express Protect Technet24

||||||||||||||||||||

||||||||||||||||||||

DNS, NTP, and other networking need to be configured correctly (as per best practices). If application-consistent snapshots are required, use the appropriate application snapin. RMC VM is hosted on either ESXi or Hyper-V. Define a Recovery Set (contains info on the virtual volume, array type, and array identifier). Create a Snapshot Set (associate it with the Recovery Set). FC will need to be zoned correctly; ports on RMC VM to be zoned exclusively to 3PAR ports. Catalyst license required on each StoreOnce used by RMC. An appropriate transport between array and RMC VM needs to be selected: SV VSA, only possible to use iSCSI 3PAR. If FC, then either VM Direct IO (ESXi) or vFC (Hyper-V), If iSCSI, then follow iSCSI best practices. An appropriate transport between RMC VM and StoreOnce needs to be selected. Support for Catalyst over Fibre Channel (CoFC)—Must be zoned correctly, a second port or set of ports on RMC VM to be zoned to the StoreOnce (Do not overlap with the RMC-> 3PAR ports!) Support for Catalyst over Ethernet (CoE)—Consider setting up a dedicated virtual NIC (vNIC) for this.

Express Restore

||||||||||||||||||||

||||||||||||||||||||

Figure 9-38 Express Restore Essentially, Express Restore is Express Protect in reverse. It recovers data at the speed of flash by sending only unique changed data from StoreOnce to 3PAR. Express Restore Features include: Deliver on deployments, consolidating a few applications with good price/performance, but limited service (SLAs) with 15X faster recovery* than traditional backup application. Reduce cost and complexity of traditional recovery approaches with direct restore from StoreOnce to 3PAR. Two Catalyst Backups are compared and the differences are written to an RW copy of a snapshot. This is extremely efficient and offers greater speed improvements over conventional methods. This is not a one-size-fits-all strategy but works well in the DevOps world. RMC performs a block comparison between Express Protect Backups. This is then used to create a “child” R/W snapshot on the 3PAR. This can then be “attached/mounted” to a host to recover files.

Element Recovery Telemetry Technet24

||||||||||||||||||||

Element Recovery Telemetry, shown in Figure 9-39, is the ability recover at the granularity of files in a VM environment.

Figure 9-39 Element Recovery Telemetry RMC VM acts as a proxy between the host being restored to and the Catalyst store containing the Express Protect Backup (or copy thereof). The host accesses the RMC VM using iSCSI (ESXi can use iSCSI Software Adapter). RMC accesses the actual backup using the transport that was in existence during the backup operation.

Guided workflow—new way to protect the data RMC offers a wizard for configuring the protection for various applications, as shown in Figure 9-40. RMC-O –

New Guided Protect wizard

–

Express Restore via RMC user interface

–

Scheduling enhancements

–

Specify start date, end date, and last day of month

||||||||||||||||||||

RMC-S

||||||||||||||||||||

–

New Guided Protect wizard

–

Scheduling enhancements

–

Specify start date, end date, and last day of month

RMC-SH –

New Guided Protect wizard

–

Catalyst Copy support

–

Express Restore via REST API

–

Scheduling enhancements

–

Specify start date, end date, and last day of month

||||||||||||||||||||

Figure 9-40 The new way—Guided workflow

RMC 4.1 has the ability to suspend schedules Before 4.1, schedules needed to be deleted and recreated. Starting from RMC 4.1, as shown in Figure 9-41, a schedule can be suspended.

Technet24

||||||||||||||||||||

Figure 9-41 RMC 4.1 has the ability to suspend schedules

RMC-V Snapshots/Veeam Explorer Integration RMC-V: Application-consistent snapshots of virtual machines –

MS SQL

–

Exchange

–

Filesystem

||||||||||||||||||||

RMC-V uses VMtools/VSS to quiesce Microsoft filesystems and applications. RMC then snapshots the entire Virtual Volume. Consider using Express Protect to take a backup of the snapshot. StoreVirtual VSA 12.6, 12.7, and HC250 supported, as well as 3PAR. Veeam: Can leverage array snapshots taken by RMC-V. Single Item recovery performed on these snaps. Restore emails, database schema, and individual documents. No specific version requirements from Veeam or RMC.

||||||||||||||||||||

Learning check Peer Copy is supported with RMC 4.1 and greater. Which features are supported? (Select 2.) A. Homogenous replication between StoreVirtual VSA and 3PAR B. Heterogeneous replication between StoreVirtual VSA and 3PAR C. Unidirectional replication D. Bidirectional replication

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Creating snapshots and Express Protect backup for Oracle databases video Instructions: Watch the “Creating snapshots and Express Protect backup for Oracle databases” video at: https://www.youtube.com/watch?v=3cfQc2wfQkI. Notes:

||||||||||||||||||||

||||||||||||||||||||

StoreOnce operating system features and benefits Hardware Appliances: StoreOnce Catalyst Copy over Fibre Channel (note VTL/NAS replication is via IP network only). Performance improvements in Dedupe “engine”. Component Firmware update process improved to reduce the system downtime. Updates on multinode systems are now performed in parallel. There is a new CLI command syntax: Reporting Central upgrade to equivalent functionality of SEM. New functionality ready for Veeam 10 (Catalyst Copy, IVRM). Static Routing can be configured for single-node systems.

StoreOnce Virtual Appliances: StoreOnce VSA now has HPE branding. Support for FC connection. New 32TB license and new capacity upgrade licenses. Available in Microsoft Azure Marketplace. Deployment in Microsoft Azure Cloud: Users can now deploy StoreOnce VSA in Microsoft Azure cloud. Multiple StoreOnce VSA instances can be deployed across Microsoft Azure cloud. As part of the right mix of on-premise and off-premise storage, data protected on HPE StoreOnce Systems can now be replicated to a StoreOnce VSA hosted in the Azure cloud. To data protection applications, the off-premise StoreOnce VSA is managed the same as other secondary storage targets. Increased efficiency and performance: StoreOnce VSA 3.16 delivers twice the backup performance of the current StoreOnce VSA with speeds up to 12TB/hr. with no increase in resources that need to be allocated to running the StoreOnce VSA.Clou Using backup data in a VSA in Azure is functionally no different to using backup data on a local StoreOnce to restore from. The difference is the bandwidth of the connection Technet24

||||||||||||||||||||

||||||||||||||||||||

from the local media server to the Azure-hosted VSA which will likely mean slow restores. As an alternative, the restore can be redirected to an Azure-hosted VM. This, of course, requires the DR infrastructure to be provisioned in Azure and still requires an eventual restore to a local target. StoreOnce Cloud Bank is a StoreOnce feature that enables StoreOnce to leverage external object storage for the long-term retention of backup copies.

HPE StoreOnce Cloud Bank Storage StoreOnce Cloud Bank, shown in Figure 9-42, leverages object storage for long-term retention of backup data copies. Features include: Control of copy to object storage using existing backup application. Ability to take advantage of StoreOnce deduplication-optimized copy to the cloud. Multithreaded upload to cloud to minimize effects of latency. Metadata for externally stored backup copies is cached locally. Support for multiple object storage vendors. Ability to leverage public cloud services for DR.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-42 HPE StoreOnce Cloud Bank Storage

HPE StoreOnce Cloud Bank Storage use cases Use Case 1—Long-Term Backup Retention Wants to retain data for a long time as cheaply as possible. Plan to seldom read the data, if ever. Use Case 2—Offsite DR Protection Wants to regularly send backups offsite for the purposes of DR protection. No secondary site to replicate too. Only plan to read the data for DR recovery.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Use Case 3—StoreOnce VSA in the Cloud backed by Object Storage Using StoreOnce VSA as a backup/replication target within Cloud Compute. Would like to reduce running costs by hosting data on cheaper object storage.

HPE StoreOnce Cloud Bank Storage requirements

Figure 9-43 HPE StoreOnce Cloud Bank Storage requirements Azure has different access tiers; Hot and Cool. A Hot access tier which indicates that the objects in the storage account will be more frequently accessed. This allows you to store data at a lower access cost. A Cool access tier which indicates that the objects in the storage account will be less frequently accessed. This allows you to store data at a lower data storage cost. Note The following link suggests that there is no performance impact of using a Cool tier over a Hot tier. Choice is driven by the financial considerations. https://docs.microsoft.com/en-us/azure/storage/storage-blob-storage-tiers Pricing information can be found here: https://azure.microsoft.com/en-us/pricing/details/storage/blobs/

||||||||||||||||||||

||||||||||||||||||||

During the controlled release, Cloud Bank capacity is limited to max local capacity less configured capacity. Maximum supported Cloud Bank capacity is shown in Figure 9-44 and Figure 9-45.

Figure 9-44 Maximum supported Cloud Bank capacity

Figure 9-45 Maximum supported Cloud Bank capacity Configuring Cloud Bank using a Catalyst store and select preferred Cloud Service provider is shown in Figure 9-46 and Figure 9-47.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-46 Configuring Cloud Bank using a Catalyst store

Figure 9-47 Select your preferred Cloud Service provider

Setting encryption If restoring a Cloud Bank store and attaching to a store that is encrypted, the user must

||||||||||||||||||||

||||||||||||||||||||

import the key, as shown in Figure 9-48, if it is not already available on the device before they can attach. The user must provide the encrypted key, and the password used to encrypt it to enable the store to reattach.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-48 Importing a key for encryption

RMC architecture with Cloud Bank attached RMC support for HPE Cloud Bank Storage lets customers leverage the economics, agility, and flexibility of the cloud for modernized data protection. Backup data can be seamlessly, securely, and cost effectively moved to the public, private, hybrid cloud, or on-premises object storage—enabling long-term retention, archive, and reliable DR that is simple and efficient. Figure 9-49 shows architecture of RMC with Cloud Bank.

Figure 9-49 RMC architecture with Cloud Bank attached Features include: Support for Hyper-V as well as VMware (only VSA) Low-bandwidth transfer to catalyst store using 16G FC or 10G Ethernet Cloud Bank uses a second Catalyst store for the cloud then Catalyst copy to the Object store in the cloud, as shown in Figure 9-49. Create Catalyst store on the StoreOnce this will use low-bandwidth transfer to catalyst store. Using RMC to do a snapshot of a VV with Express Protect backup to the Catalyst store. Use Catalyst copy to do a copy to the Cloud Bank Catalyst store which, in reality, is copying to the Object store using storage in the cloud.

||||||||||||||||||||

||||||||||||||||||||

Data Protector and object copy to Cloud Bank Features of copying object data to Cloud Bank using Data Protector, shown in Figure 950, include: The Backup server with Media agent creates a backup to the StoreOnce Catalyst store. The Disk agent pulls the data off the Server to create the backup. If using Data Protect, you must use Server side explicit gateway and dedupe locally. (This would be the same with NetBackup using OST). Metadata is saved locally also in StoreOnce for the Object store copy. Object copy does the copy to the second Catalyst store which is again, in reality, in the cloud. You must use the Explicit gateway (Server side), so the deduplicating and metadata is held locally. Use Storage Lifecycle policies with NetBackup to perform the object copy.

Figure 9-50 Data Protector and object copy to Cloud Bank

StoreOnce Cloud Bank support There are many differences between the Cloud Storage Providers, shown in Figure 951. For example: Authentication schemes used by each providers, tokens, keystone IAM, and so on.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Different APIs, support for different operations. Architectural limits of some providers. For example, Azure 500TB per storage accounts. Initial version supporting Amazon web services, Microsoft Azure and Scality. Future version targeted to support Swift and Google Cloud.

Figure 9-51 StoreOnce Cloud Bank support

Licensing

||||||||||||||||||||

Figure 9-52 Licensing Cloud Bank licensing features include: Simple licensing model. Licenses are available in 1TB increments up to two times the total licensed max capacity for the StoreOnce appliance. 3 types of license are available. Cloud Bank Storage Base read/write LTU. Cloud Bank Storage Archive option. Cloud Bank Storage Multi-Reader option. RAM upgrade kits available (for maximum Cloud Bank Storage option system RAM will need to be upgraded. Cloud Bank licensing considerations: StoreOnce 3540—Maximum local licensed capacity 31.5TB –

Maximum Cloud Bank Storage capacity (at GA) 2x31.5=63TB.

–

Customer wants full capacity for Cloud Bank Storage, Read/Write capability.

–

63x1TB Read/Write purchased.

–

1 Cloud Bank Storage license key issued to be entered on system.

–

RAM upgrade kit will also need to be added.

StoreOnce 5500—Maximum local licensed capacity 150TB –

Maximum Cloud Bank Storage capacity (at GA) 2x150=300TB.

–

Customer wants full available capacity for Cloud Bank Storage Read/Write and Archive option.

–

300x1TB Read/Write purchased.

–

300x1TB Archive option purchased.

–

Two Cloud Bank Storage license keys issued to be entered into system.

–

RAM upgrade kit not required.

||||||||||||||||||||

Reporting Central Technet24

||||||||||||||||||||

Figure 9-53 Reporting Central The aim is to implement equivalent functionality in Reporting Central (RC) of SEM. SEM will go End of Life and will be removed from the Software Depot. Two new functionalities released in RC with 3.16.2 software: Capacity threshold alerts –

Customer can automatically trigger email alerts when the appliance crosses a customized capacity threshold.

–

Capacity threshold can be configured for local and remote appliances. In case of multimode, it is possible to select individual service sets to be monitored.

–

Threshold can be configured in 10% increments stating from 45%.

–

Reporting frequency either a set interval (1/2/3/6/12/24 hours) or at 10% increase after the 45% threshold.

–

Alert type can be configured to Info, Warning, Critical. (The system defines Warning at 85% capacity used and Critical at 95% capacity used.)

Report scheduler –

Customers can schedule reports directly to their email including local and remote appliances (one service set in one report).

–

Two ways to schedule a report, through the Reports screen and the Scheduler.

Three report types are available: –

Capacity Usage

–

Read/Write Throughput

–

Replication Throughput

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Reporting period between “Last 5 minutes” and “Last 1 Year”. Report arrives in email with embedded picture as well as CSV file.

Catalyst Copy over Fibre Channel Catalyst Copy over Fibre Channel: Extends CoFC technology to move data between StoreOnce Appliances. Catalyst copy before 3.16.2 release is via Ethernet. Satisfies a customer need where they have FC-only environments or their Ethernet link is too slow. Backup applications that support Catalyst do not need to be modified for Catalyst Copy over FC. Uses Data Protector Object Copy or NetBackup Storage Lifecycle Management. (Backup Exec 2016 will be supported with the OST 4.1 plug-in later in 2017.) VTL and NAS replication are NOT supported using FC and can use only Ethernet. StoreOnce Appliance can use Catalyst Copy over Fibre Channel and Catalyst Copy over Ethernet if required. Backup can be via Ethernet and copy via FC and vice versa. Additions to the Catalyst GUI to support this feature. FC Loop support is NOT supported (that is, direct FC connection not supported). CoFC identifier is substituted for IP address/FQDN (Fully Qualified Domain Name).

Catalyst Copy over Fibre Channel—Use cases Figure 9-54, Figure 9-55, Figure 9-56, and Figure 9-57 illustrate examples of possible use cases for Catalyst Copy over Fibre Channel.

Figure 9-54 Backup over Ethernet, then copy job over Fibre Channel

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-55 Backup over Fibre Channel, then copy job over Fibre Channel

Figure 9-56 Copy over Fibre Channel, then restore over Ethernet

Figure 9-57 Copy job over FC and Ethernet simultaneously from the same source item (multireader) to different target items Figure 9-58 and Figure Figure 9-59 illustrate examples of possible use cases for Catalyst Copy over FC with fan-in and fan-out relationships.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-58 Copy job over FC and Ethernet simultaneously from the same source item (multireader) to different target items

Figure 9-59 A single-source appliance copying to multiple target StoreOnce Appliances (fan-out) over a combination of FC and Ethernet

How to zone correctly for Catalyst Copy over FC Three zones are recommended as shown in Figure 9-60, one for the actual backup and two for Catalyst Copy: ISV Media server initiator WWN (World Wide Name) should be zoned with the Target WWN on both the appliances in zone 1. The appliance 1 Target WWN should be zoned with StoreOnce appliance 2 Initiator WWN in zone 2. The appliance 2 Target WWN should be zoned with StoreOnce appliance 1 Initiator Technet24

||||||||||||||||||||

||||||||||||||||||||

WWN in zone 3.

Figure 9-60 How to zone correctly for Catalyst Copy over FC

||||||||||||||||||||

||||||||||||||||||||

Learning check The StoreOnce 3100 is supported with Cloud Bank. True False

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: HPE Cloud Bank Storage video Instructions: Watch the “HPE Cloud Bank Storage” video at: https://h20195.www2.hpe.com/V2/GetDocument.aspx?docname=A00029858ENN. Notes:

||||||||||||||||||||

||||||||||||||||||||

HPE StoreEver Tape portfolio HPE offers a broad LTO tape portfolio, as shown in Figure 9-61, including tape media, standalone tape drives, and enterprise-class tape libraries.

Figure 9-61 HPE StoreEver Tape portfolio The HPE T950 and HPE TFinity ExaScale libraries are positioned to replace the HPE ESL G3 and will comprise the Enterprise space within the HPE Tape Portfolio. The new LTO-8 specifications show an increase of two times the tape cartridge capacity from the previous LTO-7, with customers now being able to store up to 30TB* per cartridge when compressed (12TB native). Customers who need to consolidate backup to a single device, using a low-cost, longterm archiving solution, and who require more than 560 slots would benefit from these new tape library offerings. These solutions feature hardware-based encryption for secure, off-line storage. They include manageability tools such as HPE StoreEver TapeAssure Advanced Software, a software utility that proactively monitors the utilization performance and health of a customer’s tape environment. The StoreEver Tape portfolio also includes support for HPE StoreOpen with Linear Tape File System (LTFS).

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE StoreEver—LTO Tape Drive/MSL Tape Libraries The HPE StoreEver, shown in Figure 9-62, is the ideal solution for customers: Implementing their first backup and/or archive of Direct Attached Storage (DAS). Need for backup/archive data to be retained for long periods—years. Off-site backups/archives are a requirement for DR or regulatory reasons. ROBO environments. Looking for a cost-effective and efficient DR repository/archive. HPE StoreEver LTO tape drives—high capacity and low cost of ownership, proven reliability, best practice media rotation, cost-effective/durable archiving. HPE StoreEver Tape Autoloaders—the ideal solution for archiving and long-term data retention with industry-leading affordability, capacity, and security features (encryption and write once, read many [WORM]). HPE StoreEver MSL Tape library—unattended backup, DR, and archive solution with web-based remote management; the ideal solution for archive and long-term data retention due to industry-leading affordability; capacity, fast.

Figure 9-62 HPE StoreEver—LTO Tape Drive/MSL Tape Libraries

HPE LTO Tape Drive portfolio HPE StoreEver Tape drives, shown in Figure 9-63, have the lowest cost per terabyte for long-term storage with limited power or energy requirements. These drives are ideal for archiving cold or active data with a media shelf life of up to 30 years in normal, ambient conditions. You can easily transport rugged LTO cartridges off site for DR purposes and use WORM media for protection against accidental overwrite. LTO data cartridges provide portable, high-density storage.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-63 HPE LTO Tape Drive portfolio

LTO technical specifications LTO generation 8 specifications will continue to include features introduced in previous generations such as multilayer security support including hardware-based encryption and WORM functionality. Additionally, the partitioning functionality that allows users to present an easy-to-use tape-based file system with the use of LTFS will be included. Alongside the launch of LTO Ultrium generation 8 specifications, for customers who purchase new LTO generation 7 cartridges, a new initialization feature will be introduced to enable the cartridge capacity to be increased by up to 50%. New LTO generation 7 cartridges initialized as LTO-8 Type M media will be able to store up to 22.5TB* of data. Together, these features are expected to provide users with a costeffective storage solution that is both easy-to-use and addresses numerous storage needs. LTO technical specifications are shown in Figure 9-64.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-64 LTO technical specifications StoreOpen LTFS provides file management capabilities, making tape as easy to access and share as disks. LTFS enables file retrieval from long-term storage without depending on backup software or hardware. Administrators can manage, configure, and use HPE StoreEver ESL G3, MSL6480, MSL G3, and 1/8 G2 Tape Autoloader libraries from anywhere. These solutions include web-based remote management and diagnostic tools that can proactively and intelligently monitor utilization, operational performance, and overall life and health of the library, drives, and media. Note LTO-8 doubles the tape cartridge capacity compared to LTO-7.

HPE StoreEver 1/8 G2 Tape Autoloader family comparison The HPE StoreEver 1/8 G2 Tape Autoloader, shown in Figure 9-65, provides a costeffective and easy-to-manage tape solution combining incredible storage density and features into a compact 1U form factor.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-65 HPE StoreEver 1/8 G2 Tape Autoloader family comparison LTO-8 Ultrium 30750 tape technology delivers 240TB of compressed capacity (2.5:1) in a 1U form factor. Easy to install, manage, and use integrated solution.

HPE StoreEver MSL 6480 comparison HPE StoreEver MSL6480 tape library, shown in Figure 9-66, provides midrange tape automation, delivering scalability, density, and performance to meet customers’ shortterm backup, DR data protection, and long-term archival requirements. Start with an 80slot 6U form-factor Base Library Module, and add up to 6 Expansion Library Modules for a total of 560 slots in a 42U rack form factor, 16.8PB of total maximum capacity with 560 slots and LTO-8 drives.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-66 HPE StoreEver MSL 6480 comparison Performance scaling from 1 to 42 LTO half height (HH) Tape Drives and native transfer rates up to 300 MB/s per LTO-8 Tape Drive.

HPE TFinity ExaScale Edition The HPE T950 Tape Library combines field-proven reliability with storage innovation and gives you access to a full suite of enterprise storage capabilities. It delivers powerful library management, superior and proactive data integrity administration, and delivers outstanding TCO of any tape solution. Choose from a range of base configurations and scale up to 53,460 LTO-8 tape cartridges or 40,680 TS11xx Technology cartridges in increments of 10 LTO of 9 TS11xx Technology slots for capacity on demand. Incorporating between 1 and 144 tape drives, you can consolidate and store up to 1.52 EB (compressed 2.5:1) using TS11xx Technology or 1.6 EB (compressed 2.5:1) using LTO-8 tape cartridges of enterprise data. HPE’s TFinity ExaScale Edition tape library, shown in Figure 9-67, provides maximum flexibility to use either LTO-6/LTO-7/LTO-8 or TS11x0 tape technology enabling you to choose the right media for your business needs.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-67 HPE TFinity ExaScale Edition As with the HPE T950, the HPE TFinity ExaScale Edition tape library is built with the highest performance achievable in automated tape technology. From robotics to drives, to software, and to media, all performance innovations are included in these libraries. When data center real estate counts, HPE’s TFinity ExaScale Edition offers you unsurpassed storage density and the smallest footprint through a unique and highly efficient library design. More storage capacity in less floor space. Using innovative “shelves” instead of slots and TeraPack® containers in place of individual cartridges, HPE’s TFinity ExaScale Edition industry-best density delivers up to a 50% reduction in data center floor space required versus competing offerings.

Technet24

||||||||||||||||||||

||||||||||||||||||||

BlueScale tape library software Every HPE T950 tape library has BlueScale® software interface built-in, so customers have intelligent command, control, and data integrity functionality at the touch of a screen. A wide range of Spectra support options and professional services are also available to customers including Assisted Self-Maintenance, global spare drive options, and installation and integration services. Managing HPE TFinity ExaScale tape Library with BlueScale Universal Interface, shown in Figure 9-68, allows you to use: Configuration Remote Management Shared Library Services Encryption Key Management Media Lifecycle Management Drive Lifecycle Management Library Lifecycle Management Auto Support

||||||||||||||||||||

||||||||||||||||||||

Figure 9-68 BlueScale tape library software

New HPE T950V with LTO-8 The HPE T950V, shown in Figure 9-69, is a lower-cost HPE T950 variant that will be available for price-sensitive customers. The HPE T950V will be similar to the existing HPE T950 tape library offering today, but with a set of product differences listed on the screen as a trade-off for the lower priced configuration.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-69 New HPE T950V with LTO-8 The downsides will vary by item for each customer, but in most cases, there is little to no impact. Customers will find that the differences are often only less convenient or cosmetic. However, the lack of a back door will be an issue for many customers to be aware of. As it might cause inconsistencies with a “clean and tidy” data centers, it might create a perception of the tape library being less secure. Or there might be perceptions related to the risk of snagging on cables and inferior air flow cooling and shielding from electromagnetic interference. However, additional testing has been completed to negate this last point (testing on cooling and electromagnetic interference, however, has been completed). HPE T950V specifications are shown in Figure 9-70.

Figure 9-70 HPE T950V specifications

||||||||||||||||||||

||||||||||||||||||||

HPE T950V specifications Customer is willing to accept the product differences in favor of some additional cost savings, including lower-cost SKUs, across the board Spectra reductions and additional supply chain savings, customers could benefit from approximate savings of 9%–10% on their overall configuration.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check Existing MSL Tape Drives and Tape Media can be used in the MSL6480. True False

||||||||||||||||||||

||||||||||||||||||||

HPE tape tools StoreEver TapeAssure Advanced Software and CV-TL StoreEver TapeAssure Advanced Software provides an intuitive, easy-to-use dashboard, along with advanced analytics to support the performance of tape drives, libraries, and cartridges. StoreEver TapeAssure Advanced Software is fully integrated into HPE StoreEver Command View Tape Library (CV-TL) Software, as shown in Figure 9-71. This provides a single view of management, multilibrary manageability, monitoring, usage, performance, and analytics.

Figure 9-71 StoreEver TapeAssure Advanced Software and CV-TL HPE tape libraries and software solutions offer a simple dashboard to quickly identify all aspects of the library and its components. The basic functionality of CV-TL is included with every HPE StoreEver MSL or ESL tape library. However, StoreEver TapeAssure Advanced Software is a licensed feature of CV-TL that captures and analyzes an unparalleled quality and quantity of health metrics to proactively and predictively monitor the performance, health, and utilization of every StoreEver tape cartridge, drive, and library in the infrastructure.

Technet24

||||||||||||||||||||

||||||||||||||||||||

StoreEver tape libraries and manageability software StoreEver ESL G3 and StoreEver MSL6480 tape libraries are designed to ease the management burdens of data protection in the SAN. StoreEver tape libraries and HPE Command View for Tape Libraries Software makes managing data more efficient by delivering a visual dashboard that quickly identifies all aspects of each tape library in the environment.

HPE Library and Tape Tools HPE Library and Tape Tools (L&TT) is a free diagnostic tool for all HPE tape storage and magneto-optical storage products. Targeted for a wide range of users, L&TT is ideal for customers who want to verify their installation, ensure product reliability, perform their own diagnostics, and achieve faster resolution of tape device issues: Actively measure the health of your HPE tape hardware. Get the most out of your device through access to operational, health, and usage data in the support ticket. Reduce product downtime through firmware updates, preventative maintenance, and fast issue diagnosis with corrective actions. Access to comprehensive device analysis and diagnostic tests. First-level failure analysis of both the device and system without HPE involvement. Troubleshoot system performance issues. Identify the most common host system configuration issues. Easy access to any advisories for your standalone drives and libraries. Easy to use and best in class support tool as used by the HPE hardware support organization. Generate and email comprehensive support tickets to the support center for faster troubleshooting. Free download and install from HPE.com (http://www.hpe.com/support/tapetools) in less than five minutes. Can run directly from a memory stick without installing on your production server. L&TT performs firmware upgrades, verification of device operation, failure analysis, and a range of utility functions. Performance tools assist with troubleshooting bottlenecks and system configuration checks alert users of common host issues.

||||||||||||||||||||

||||||||||||||||||||

It also provides seamless integration with HPE support by generating and emailing test results and support tickets. HPE requires the use of L&TT to troubleshoot most device issues, so it is recommended that customers initiate a support ticket and run the device assessment test before calling.

Ease of use with HPE StoreOpen and LTFS

Figure 9-72 Ease of use with HPE StoreOpen and LTFS LTFS is a file system for tapes: Presents LTO tape as another drive letter, icon, or folder like a disk or memory stick. Manages content easily with drag and drop directory tree access. Uses LTO-5, LTO-6, LTO-7, or LTO-8 dual-partitioning capability to store the index and the content. Self-describing tape: No application software dependencies Cross-platform sharing of content

Technet24

||||||||||||||||||||

||||||||||||||||||||

Ideal for managing long-term archives HPE StoreOpen and LTFS are available as a free download for HPE StoreEver LTO-8 Standalone Tape Drives and HPE StoreEver 1/8 G2 Autoloader customers. Note Learn more at www.hpe.com/storage/ltfs

What is NinjaProtected+? NinjaProtected+ is an HPE-developed replacement for a third-party backup assessment tool that has been in use for many years. Get a free noninvasive backup storage assessment from HPE and receive a detailed report including: Current backup capacity and performance Key performance indicators such as backup windows Backup success rates Opportunities for cost reduction Applies to any new HPE StoreOnce System including HPE StoreOnce VSA and data types include files servers, Microsoft® Exchange, and virtual machines (using agentless backup methods). By performing a free NinjaProtected backup assessment, HPE and its partners can demonstrate how customers can reduce the amount of backup data they need to store by 95%—guaranteed (HPE StoreOnce Get Protected Guarantee Program). NinjaProtected+, illustrated by Figure 9-73, is an enterprise backup environment discovery and assessment tool that supports Data Protector, TSM, NetBackup, Commvault, BackupExec, and Networker. Using simple scripts, it collects backup environment metadata from a customer site.

||||||||||||||||||||

||||||||||||||||||||

Figure 9-73 What is NinjaProtected+? It generates reports, charts, tables, customer discussion items, storage sizing information, suggested sales opportunities, and customer workshop guides. What is new? A from-the-ground-up new development of the NinjaProtected assessment tool incl. the data collection scripts. The ability to directly access the UI to upload data and generate reports (HPE internal only in first release). Integration into Storage Assessment Foundry (SAF). HPE intellectual property model allows more enhancements in the future. Full access to anyone within HPE using AD credentials. Industry-vertical attribute to allow peer group comparisons.

Storage Assessment Foundry

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-74 Storage Assessment Foundry NinjaProtected+ is part of the Storage Assessment Foundry (SAF). SAF is a unification, reimplementation, and merge of a number of related storage assessment, sizing, and reporting tools. The integrated tools are easier to locate, will have a common look and feel, will work together, and will have real backend data integration. The Opportunities tab, shown in Figure 9-75, is the tools suggestions of items to discuss with the customer, any upsell potential and anomalies in the backup environment are listed. It is intended to allow a non-BURA specialist to quickly key into customer issues.

Figure 9-75 Looking at a customer–Opportunities tab Currently, there are four reports: Two intended to go to the customer, Executive and TCO/ROI report. One intended to be supporting material for the sales folks. One intended to be uploaded into the HPE Storage Sizer to automatically generate a StoreOnce configuration. The Executive Report, shown in Figure 9-76, is a 15-page report intended for the customer. The report contains statistics, information about the backup environment, and

||||||||||||||||||||

||||||||||||||||||||

StoreOnce and RMC positioning information.

Figure 9-76 Executive Report

HPE Storage Sizer input files On the HPE Storage Sizer, on the Backup Calculator tab, use the Import Data from NinjaProtected Tool, as shown in Figure 9-77. The Solve button is used to generate a correct StoreOnce configuration, as shown in Figure 9-78.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 9-77 HPE Storage Sizer input files

||||||||||||||||||||

||||||||||||||||||||

Figure 9-78 HPE Storage Sizer Output

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE Switch Selector tool For Ethernet and TCP/IP-based storage solutions, adequate switch might be needed. HPE Switch Selector Tool, shown in Figure 9-79, can help architects to narrow product family selections based on network needs, infrastructure requirements—or competitive equivalents. Note To access HPE Switch Selector tool, go http://h17007.www1.hpe.com/us/en/networking/products/switches/switchselector.aspx#.Wgr01ppLGTM

Figure 9-79 HPE Switch Selector tool

to

||||||||||||||||||||

||||||||||||||||||||

Learning check The HPE StoreOnce Get Protected Guarantee program is for StoreOnce systems, but not StoreOnce VSA. True False

Technet24

||||||||||||||||||||

Activity: Hewlett Packard Enterprise Switch Selector Instructions: Navigate to the "Hewlett Packard Enterprise Switch Selector" website at: http://h17007.www1.hpe.com/us/en/networking/products/switches/switchselector.aspx#.Wgr01ppLGTM. Using the switch selector tool, find a switch with the following characteristics: –

32 ports

–

Fully managed

–

Layer 3 advanced

–

40GbE port speed

–

1GbE copper uplink

–

With redundant fans and power supplies (PSUs)

Answer the questions below: –

How much flash?

________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ –

How much DRAM?

________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ –

What is the 10 Gbps latency?

________________________________________________________________________ ________________________________________________________________________ ________________________________________________________________________ –

||||||||||||||||||||

What is the throughput?

________________________________________________________________________

||||||||||||||||||||

||||||||||||||||||||

________________________________________________________________________ ________________________________________________________________________ Note Hint—look under “Product Details.”

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check Write a summary of the key concepts presented in this chapter.

||||||||||||||||||||

||||||||||||||||||||

Summary HPE StoreOnce products support entry-level to high-end backup and recovery through a simple standardized interface. HPE StoreOnce provides a simple, consistent, high-performing architecture that spans the organization. HPE StoreOnce RMC serves as the single integration and control point for data protection. HPE StoreOnce Cloud Bank Storage leverages object storage for long-term retention of backup data copies. HPE StoreEver Tape portfolio includes tape media, standalone tape drives, and enterprise-class tape libraries.

Technet24

||||||||||||||||||||

||||||||||||||||||||

10 Designing HPE BURA Solutions

||||||||||||||||||||

||||||||||||||||||||

LEARNING OBJECTIVES After completing this chapter, you should be able to: Describe availability and downtime. Identify data tiering and retention. Describe archiving solutions from Hewlett Packard Enterprise (HPE). Describe backup strategy planning. Discuss the features of software used to support HPE backup, recovery, and archive (BURA) solutions.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Prelearning check Match each severity level with its definition in Figure 10-1.

Figure 10-1 Prelearning check

||||||||||||||||||||

||||||||||||||||||||

Availability and downtime Downtime of an IT system can cause a loss of productivity for a single user, workgroup, or even the entire company. Whenever downtime impairs business, the outage carries serious consequences. Quantifying downtime is not an easy task because downtime differs from one case to another. Losing a second in an air traffic control system or in a hospital life-support system might have a critical impact, whereas losing hours in billing system might not have a significant impact at all, since transactions are queued and committed once the system becomes available. There are causes of downtime which cannot be predicted (such as natural disasters or fire), but the failure rates of a computer system or its component can be predicted with a level of certainty. To determine this, the manufacturer provides a Mean Time Between Failure value (MTBF). Definition: Failure is the termination of the product ability a whole to perform its required function. MTBF is an interval of time used to express the expected failure rate of a given product. It does not indicate the expected lifetime of that product and says nothing about the failure likelihood of a single unit. A disaster in the IT world is any event that interrupts a company’s computer operations (and thus causes downtime). Disasters are not controllable, but precautions against disasters make the recovery from such events easier. Disasters are categorized according to the affected area: Building-level incidents Disasters affecting a building usually influence computer operations in that building as well. There might not be a direct damage to the system, but these incidents might prevent access to systems or interrupt operations. Metropolitan area disasters Usually floods, fires, large chemical incidents, moderate earthquakes, severe winter storms, or blackouts affect entire cities, influencing IT systems. Regional events Computer operations might be interrupted by natural disasters that affect an entire region within a radius of hundreds to tens of thousands of miles/kilometers (such as large Technet24

||||||||||||||||||||

||||||||||||||||||||

floods, hurricanes, earthquakes, wars, etc.).

Planned downtime Regardless of the cause, location, or the level of downtime, administrative intervention is required to quickly recover from the problem. This intervention always depends on the exact cause of the problem and differs from one case to another. Some questions you might ask while preparing the administrative intervention are: Is it needed to take the system down for upgrades or tuning? What can you do about virus penetration or intruder/cyber-attacks? Do you have the right applications and tools? The system might be down for regularly scheduled tasks, such as database backups, index builds, table reorganizations, system maintenance, application or operating system updates, and other reasons. The majority of system and data unavailability is because of planned downtime that occurs due to required maintenance. Although unplanned downtime accounts for only about 10% of all downtime, its unexpected nature means that any single downtime incident might be more damaging to the enterprise, physically and financially, than many occurrences of planned downtime. Understanding your cost of downtime is, therefore, critical in either case.

||||||||||||||||||||

Activity: Identifying causes of downtime Instructions: Here are some causes of system downtime leading to data unavailability: –

Power outages

–

Data center cooling issues

–

Viruses

–

Cyber/intruder attacks

||||||||||||||||||||

Name three more causes of system downtime.

What is backup?

Technet24

||||||||||||||||||||

||||||||||||||||||||

First, consider these questions and comments regarding backup: Is backup a protection against hardware failures? Backup does not fix disk problems! Is backup a protection against loss of data due to hardware failure? Backup does not always have the latest data. Therefore, even with the best backup solution, there is some probability of data loss. In most cases, the source is data on a disk, such as files, directories, databases, and applications. If the backup is expected to be used for disaster recovery, it must have data consistency. Inconsistent files could mean a serious threat to any data set when restored from a tape. Incompletely backed up files (open files with ongoing write operations while the backup software is copying them to tape) might contain inconsistent data that could be spread across various databases in the enterprise, depending on how the databases are linked to each other. Software that copies data to its destination is a backup application. The destination is a backup device such as a tape drive, with a medium to which the copy of the data is written. The backup device can be physical or virtual. There are many benefits of virtualizing the target backup devices, one of the most obvious being the potential to take advantage of the deduplication functionality (deduplication is discussed later). The second important reason for virtualizing specifically the tape devices is to overcome the limitation of sequential physical devices, such as speed and size.

What is restore? Backup is the process of copying data from one location to another for the purpose of preserving the data. A critical component of any backup is the backup strategy. A backup strategy is needed to minimize the most common consequences of data loss– related problems. A restore, shown in Figure 10-2, is a process that recreates the original data from a backup copy. This process consists of the preparation and actual restore of data, and some post-restore actions that make that data ready for use.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-2 Restore process The source of the restore process is the backup copy. A restore application is the software that writes data to its destination. The destination is usually a disk to which the original data is written. When backing up the environment, consideration should be given to what is to be backed up and to where. Some of the backup preparation steps include: Identifying the data that must to be saved, Grouping data into “jobs” Assigning importance levels to this data, Calculating the backup frequency Protecting the data, Testing the backup Maintaining backup history, Archiving the tapes Defining the restore procedures

Identifying the data to be saved A typical company’s data doubles, even triples, every year! What must be backed up? All data that a business relies on should be backed up, including the operating systems, applications, and most importantly, user data. The entire system must be restorable. Many people consider the operating system not worth being backed up because there are distribution media available to restore crashed systems. However, the enormous work that was put into installing and configuring the system is then lost.

Job grouping Decide how to best group the data by considering the level of importance, the volume of data, and the recovery requirements. You can group the backup jobs based on:

Technet24

||||||||||||||||||||

||||||||||||||||||||

Devices Directories Media sets Partitions Workgroups

Backup frequency Some factors that can help decide the backup frequency include: Rate at which data is created/changed in the company Acceptable data loss (recovery point) Value of the data Cost and time to save the data Allowed downtime to restore the data While considering your design you will also need to determine the type of backup, is it to be a full backup, copying all the data, or maybe a differential or Incremental backup. Will it be completed Online or Offline or do we need an image backup? The amount of data and the time it takes to do your backup might influence this decision.

Differential backup Each type of backup, shown in Figure 10-3, has its pros and cons that users can choose from to decide what best fits their business needs. The truth about data backup is that each situation should have a custom data backup plan.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-3 Backup types A differential backup, shown in Figure 10-4, copies all files that have been modified or created since the last full backup but does not modify the archive bit.

Figure 10-4 Differential backup This is useful when you want to have the latest version of files on a tape. When the same tape medium is used for the differential backups between full backups, usually the newer file versions are allowed to overwrite the older versions of the same file. Special caution must be taken when database log files are backed up—circular logging must be disabled, and the log files must not be deleted after the backup. Since the archive bit of the files does not change during the differential backup, the files appear to the system as not being backed up and are saved again with the next differential backup (some of the same data might have been changed again, however). The advantage of the differential backup is that only two tapes are necessary to completely restore the system. The disadvantages are the volume of data on the tapes, which grows every day, the time for the backup, which also increases every day, and only the latest versions of a file can be restored.

Technet24

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Incremental backup An incremental backup, shown in Figure 10-5, copies all files that have been modified or created since the last full or incremental backup and turns the archive bit off.

Figure 10-5 Incremental backup All files that have been created or modified since the last full or incremental backup are saved. Each revision or version of a file is available on the tape because the software does not overwrite the earlier versions of the same file but appends the data to the previous backup sets (or uses different tapes for each incremental backup). When backing up database log files, circular logging must be disabled. Like normal backup, the incremental backup purges the log files after the backup process. The advantages of the incremental backup are that the size of backed up data on each tape remains almost the same and the backup window is small each day. The disadvantage is that each day must be restored separately, consuming time.

When to choose incremental vs. differential backup Technet24

||||||||||||||||||||

||||||||||||||||||||

Choose incremental backup when: You have a high percentage of daily changes in your data set. You only have a limited backup window per day. You can accept a relatively slow restore (long downtime). Choose differential backup when: You have a low percentage of daily changes in your data set or the same files change every day. The backup window is not an issue (you have enough time available for the backup). You need a relatively fast restore (short downtime). Selecting the appropriate type of backup (incremental or differential) is a trade-off between the time needed for the backup and the time needed for the restore, which depends on the size of data and the daily change rate. The type of backup must be considered when selecting the right tape rotation scheme.

Synthetic backup During the backup process, data is moved from many clients to the central backup server and then moved to a permanent storage such as a tape library. Moving data around the data center is always expensive, and any operation such as creating a full backup of many clients can overload the network infrastructure. Considering that all backup sets (full, incremental, and differential) end up on a single backup server, it is possible to take the last full backup and append the incremental or differential backups to the full backup set. Once the appending procedure is done, the result is stored to the same backup target as the source data sets and is promoted to the full backup. This methodology is called synthetic backup. The main benefit of synthetic backups over classic full backups is that it does not involve clients or moving the information over the network. If planned properly and executed at the right backup server at the right time, synthetic backup can be created almost instantly. RMC uses Inline synthetic backup to copy data to HPE StoreOnce.

Conceptual illustration of RMC Express Protect—Inline

||||||||||||||||||||

||||||||||||||||||||

synthetic full Figure 10-6 represents how inline synthetic full backup is being synthetized on the block level.

Figure 10-6 Synthetic full on the block level The backup is always a full backup; in our example, this copies B0 – B9 to C0 – C9 on the StoreOnce array. A manifest file-1 is created, identifying the data copied to the StoreOnce array. The second backup now synthesizes through a Snapdiff data B11, B12, and B5. This data is then copied to the StoreOnce array updating the manifest file for C11 and C12. However, as B5 is duplicative data, this is not copied, and a pointed is used to identify its location through the manifest file-2.

Archive bit When a file is created or changed, the operating system maintains a flag, which is called the archive bit. The backup software uses this bit, shown in Figure 10-7, to determine whether a file has been backed up before. As soon as a file is backed up using either the full or incremental backup, this bit is turned off, indicating to the system that the file was saved. If the file is changed again, the bit is turned back on and the file is flagged to be

Technet24

||||||||||||||||||||

||||||||||||||||||||

backed up again by the next full or incremental backup. Differential backups include only files that were created or modified since the last full backup. When a differential backup is performed, no changes are made to the archive bit.

Figure 10-7 Archive bit

Maintaining history A backup strategy should attempt to: Maintain a file history by saving several versions over a period of time Minimize the backup window by only saving what has changed The solution for these tasks is a tape rotation scheme. In addition to the standard features, the tape rotation schemes provide older copies of data on tape, which can be archived off-site in a secure location to provide disaster tolerance.

Backup tape rotation schemes A good plan is necessary to ensure that your backups are performed at the appropriate intervals and not only when needed—because data is valuable and difficult to replace. Most companies back up their data on a daily basis when the network is least busy. These backups might need to be performed more or less frequently, based on the criticality of the data. A regular and scheduled backup should address these issues: Speed of recovery after a disaster occurs Redundancy and history of the data Efficient use of tape media (although tapes are relatively inexpensive) Automation helps to increase the overall speed of both backup and restore. The appropriate backup schedule and the data version history can be automated with a tape rotation scheme. The longer a company needs to keep its data, the more portable media (tape cartridges) are needed. The three biggest advantages of tape rotation schemes are: Automation Archiving

||||||||||||||||||||

||||||||||||||||||||

File history Working with tape drives requires planning. Considering a relatively limited capacity and a high price of the tape media compared to the volume of data being backed up, you want to reuse tape cartridges by rotating them. While using a single external tape drive, it is the administrator’s task to replace the tapes and enforce the rotation order. Modern high-capacity tape libraries implement the tape rotation themselves, and the administrator selects the rotation scheme as an option within the backup server settings. The most-popular backup tape rotation schemes are: First In, First Out Grandfather-father-son A First In, First Out (FIFO) backup scheme saves new or modified files on the oldest medium in the set (for example, the medium that contains the oldest and thus least useful previously backed up data). Performing a daily backup onto a set of 14 media, the backup depth would be 14 days. Each day, the oldest medium would be inserted when performing the backup. The grandfather-father-son backup refers to a common rotation scheme for backup media. In this scheme, there are three or more backup sets, such as daily, weekly, and monthly. The daily backups are rotated on a daily basis using a FIFO system as above. The weekly backups are similarly rotated on a weekly basis, and the monthly backup on a monthly basis.

RPO, RTO, and retention A customer’s recovery requirements determine the best way to back up their data, and each case requires a unique strategy. Factors to consider, shown in Figure 10-8 and Figure Figure 10-9, when designing a BURA solution include: Recovery Point Objective (RPO)—How much data is allowed to go unprotected and how far back in time data must be recovered. RPO can be measured in days, hours, or even minutes. Recovery Time Objective (RTO)—How long the customer is willing to wait for the data to be recovered and the maximum allowable downtime. Retention—How long the data needs to be kept available. Retention time can range from seconds to decades, depending on company policies and government regulations.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-8 RPO, RTO, and retention

Figure 10-9 A company’s business continuity plan should indicate the RTO and RPO for business processes, such as running payroll, generating orders, and so on. The metrics specified for the business processes must then be mapped to the underlying IT systems and

||||||||||||||||||||

||||||||||||||||||||

infrastructure that support those processes. After the RTO and RPO metrics have been mapped to IT infrastructure, you can determine the most suitable strategy for each system. However, the business ultimately sets the IT budget, and therefore, the RTO and RPO metrics need to fit with the available budget. Most business unit heads want zero data loss and zero time loss, but the cost associated with that level of protection might make the desired high-availability solutions impractical.

Differences between a backup and archiving Depending on the goals of your data protection, you can decide to use archiving or a backup. Although sometimes used together in the same context, backup and archiving have a different purpose. Backup is used to keep copies of data for data protection purposes, whereas archiving is done as a means of data management for keeping the data organized for a long term. In other words, backup is used for the short-term data protection, and it might contain multiple instances of the data, whereas archiving includes arranging and indexing the information to preserve the data for a long time. Archives usually contain a single instance of the data. You can delete the original data once it is archived because accessing this information immediately is usually not required any longer. However, in reality, backed up data is usually not deleted and often continues occupying the primary storage. Backup is used for the short-term data protection, and it might contain multiple instances of the data. Archiving includes arranging and indexing the information to preserve the data for a long time. Archives usually contain a single instance of the data. You can delete the original data once it is archived because accessing this information immediately is usually not required any longer. However, in reality, backed up data is usually not deleted and often continues occupying the primary storage. Sarbanes-Oxley and HIPAA share a common theme for storage administrators: That data must be controlled and protected. There are three areas that must be examined to secure storage: Access to local. Data Access to backup data. Ensuring that backups of critical data are maintained.

Technet24

||||||||||||||||||||

||||||||||||||||||||

•

Note HIPAA is the acronym for the Health Insurance Portability and Accountability Act that was passed by Congress in 1996. HIPAA does the following: Provides the ability to transfer and continue health insurance coverage for millions of American workers and their families when they change or lose their jobs;

•

Reduces healthcare fraud and abuse;

•

Mandates industry-wide standards for healthcare information on electronic billing and other processes; and

•

Requires the protection and confidential handling of protected health information. (March 2018, http://www.dhcs.ca.gov)

Tiered data retention for HPE StoreServ Figure 10-10 illustrates a SAN that combines HPE StoreServ storage, HPE StoreOnce backup, and HPE StoreEver tape to offer a complete business protection solution. Users can implement a disk-based solution with the HPE StoreOnce backup for daily, shortterm backups and fast recovery.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-10 Tiered data retention for HPE StoreServ HPE StoreEver tape is used in conjunction with the HPE StoreOnce backup to address long-term archival and compliance objectives. As the data on the HPE StoreOnce backup appliance becomes infrequently or never accessed, the tiered data retention policies can be used to expire the data on the HPE StoreOnce while moving that data to a secondary copy on the HPE StoreEver tape. You can also transport a secondary LTO cartridge off-site for disaster recovery purposes. This reduces risk, increases efficiency, and lowers storage costs, which enable you to maximize the value you get from your data over its entire life cycle while minimizing TCO. HPE StoreEver tape is used in conjunction with the HPE StoreOnce backup to address long-term archival and compliance objectives. As the data on the HPE StoreOnce backup appliance becomes infrequently or never accessed, the tiered data retention

Technet24

||||||||||||||||||||

||||||||||||||||||||

policies, illustrated by Figure 10-11, can be used to expire the data on the HPE StoreOnce while moving that data to a secondary copy on the HPE StoreEver tape.

Figure 10-11 You can also transport a secondary LTO cartridge off-site for disaster recovery purposes.

||||||||||||||||||||

Learning check Match each severity level with its definition in Figure 10-12.

Figure 10-12 Learning check Which backup model turns off the archive bit? (Select two.) A. Full backup B. Differential backup C. Incremental backup

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Activity: Calculating the cost of downtime Instructions: Read the scenario and answer the question. Scenario: A small telecommunications sales company employing 15 sales agents each processing on average 50 orders a day. An average order has a value of $80.00. The company’s sales ordering operations run 24 hours per day. Answer the following question: What does one hour of downtime cost this company?

||||||||||||||||||||

||||||||||||||||||||

Backup strategy planning Backup strategy planning, as shown in Figure 10-13, includes: Define the requirements and constraints for backups—Determine how often data needs to be backed up or whether additional copies of the backed up data on additional media sets are needed. Understand the factors that influence the backup solution—Identify the sustained data transfer rates of the network and backup devices. These factors can affect whether the backup is network or direct. For instance, when customers back up to disk, they can take advantage of advanced backup strategies such as synthetic backup and disk staging. Prepare the backup design—Gather all the relevant information needed to design the most effective backup and recovery architecture.

Figure 10-13 Backup strategy planning Backup is really an insurance policy against data loss. A company depends on its information to stay in business, including email messages, customer databases, research and development information, or even unstructured data. Recovery is the main reason that most companies perform backups. They need to be able to: Restore a single file that has been lost or damaged Recover all data lost in a site-wide disaster

Technet24

||||||||||||||||||||

||||||||||||||||||||

Roll back to older versions of files Businesses also perform backups to comply with corporate guidelines or local legal obligations for data retention, for instance, Sarbanes-Oxley regulatory compliance. Disaster recovery consists of the processes, policies, and procedures related to preparing for recovery or continuation of the critical technology infrastructure after a natural or human-induced disaster. Disaster recovery is a subset of business continuity. Business continuity means planning for keeping all aspects of a business functioning in the midst of disruptive events, and disaster recovery focuses on the IT systems that support business functions. Archive is the migration of data from primary storage to a lower-cost long-term tier. Some estimates suggest that more than 95% of data is never accessed again 90 days after creation. Best practices for data lifecycle management recommend moving older and less-frequently accessed data to more economical platforms, such as tape. HPE provides BURA solutions to suit a variety of enterprise requirements: Backup modernization for customers struggling with antiquated backup plans and infrastructure Strategies for every stage of the planning and implementation of business protection plans Support for all major operating systems

The data protection continuum There is a range of technologies associated with data protection. Each offers a different set of benefits positioned along a continuum of performance and the risks that they address. In general, the faster the recovery time and the shorter the recovery point, the more expensive the solution. These technologies, shown in Figure 10-14, include: Tape backup—Use for lower-cost, longer-term storage of older data where time to restore is less of a factor Disk-based backup with deduplication—Use for high speed daily backup and restore, plus the ability to cost effectively retain data on disk for longer Clustering, mirroring, snaps, full system redundancy—Use for high-availability, business continuity solutions

||||||||||||||||||||

||||||||||||||||||||

Figure 10-14 The data protection continuum Customers can choose to blend storage technologies in a tiered storage solution to balance meeting RPO/RTO with cost and service-level agreements (SLAs).

Defining the requirements of a backup and recovery strategy Defining objectives and constraints of a backup strategy includes answering questions such as: Does the customer have policies about backups and restores?—Some organizations have already defined policies on archiving and storing data. A backup strategy should comply with these policies. What types of data need to be backed up?—List all types of data in the network, such as user files, system files, web servers, and large relational databases. A company’s data can be categorized as company business data, company resource data, project data, and personal data, each with its own specific requirements. How long can a backup and restore take?—Estimate the time needed for each backup and an individual or disaster recovery restore. This directly affects the amount of time data is available for use. User files can be backed up at any time when users are not working on them; however, some transactional databases might only have a few hours

Technet24

||||||||||||||||||||

||||||||||||||||||||

available for backup. The time needed for backup and restore depends on whether the type of backup is full or incremental. By backing up to disk, customers can take advantage of synthetic backup and disk staging. These advanced backup strategies significantly reduce the time needed for backup. When there is a very fast and large disk to be backed up on a slower device, consider the possibility of backing up one hard disk through multiple concurrent disk agents. Starting multiple disk agents on the same disk speeds up the backup performance considerably. What is the maximum downtime for recovery?—The allowed downtime has a significant impact on the investments into network infrastructure and equipment needed for backups. For each type of data, list the maximum acceptable downtime for recovery. In other words, determine how long specific data can be unavailable before recovered from a backup. For example, user files might be restored in two days, but some business data in large databases would need to be recovered in two hours. Recovery time consists mainly of the time needed to access the media and restore data to disks. A full system recovery takes more time because some additional steps are required. How often does data need to be backed up?—For each type of data, list how often the data needs to be backed up. For example, user working files might be backed up on a daily basis, system data on a weekly basis, and some database transactions twice a day. There are a number of factors that influence how the backup strategy is implemented. Understand these factors before preparing the backup strategy. How long should specific types of data be kept?—For each type of data, list how long the data must be kept. For example, user files might only need to be kept for three weeks, but information about company employees might be kept for five years. How should media with backup data be stored and maintained?—For each type of data, list how long the media with data must be kept in a vault, a safe, or an external location. For example, user files might not be stored in a vault at all, but order information might be kept for five years, with verification of each medium after two years. How many media sets should the data be written to during backup?—Consider writing critical data to several media sets during backup to improve the fault tolerance of such backups or to enable multisite vaulting. Object mirroring increases the time needed for backup. How much data needs to be backed up?—List the estimated amount of data to be backed up for each type of data. This influences the time needed for backup and helps customers choose the right backup devices and media for backup. What is the projected future growth of the amount of data?—Estimate future growth for each type of data. This will help ensure that backup solutions will not be quickly outdated. For example, if a company plans to hire 100 employees, the amount of user

||||||||||||||||||||

||||||||||||||||||||

and client system data will grow accordingly.

Understanding the factors that influence the backup solution The infrastructure has a high impact on backup and restore performance. The most important aspects are the parallelism of data paths and the use of high-speed equipment. Device types and models impact performance because of the sustained speed at which devices can write data to a tape (or read data from it). Data transfer rates also depend on the use of hardware compression. The achievable compression ratio depends on the nature of the data being backed up. In most cases, using high-speed devices with hardware compression improves performance. This is true, however, only if the devices stream. At the beginning and end of a backup session, backup devices require some time for operations such as rewinding, mounting, or dismounting media. Libraries offer additional advantages because of their fast and automated access to a large number of media types. At backup time, loading new or reusable media is needed. At restore time, media sets containing the data to be restored need to be accessed quickly. Data in disk-based devices is accessed faster than that in conventional devices because there is no need to load and unload media. This reduces the amount of time spent for backup and restore. Additionally, disk-based devices enable the use of advanced backup strategies such as synthetic backup and disk staging, which also reduce the backup and restore time. Data protection and archiving software focuses on using an automated Linear TapeOpen (LTO) Ultrium tape library and disk-based virtual tape backup solutions.

Storage snapshot backup considerations Consider the following when creating storage snapshot backups: HPE recommends using a combination of weekly full online backup with daily storage snapshot backups. While a snapshot is being created, no complete data backup can be performed. Storage snapshots are available only in the HPE 3PAR storage system and are not replicated to another location or other form of media. The storage snapshots cannot be used as a backup in case of a disaster from the storage system itself. Storage snapshots are part of the Express Protect backup process and are backed up to an HPE StoreOnce appliance. Volume snapshot recovery is the process of recovering from a storage snapshot and is

Technet24

||||||||||||||||||||

||||||||||||||||||||

near instant. The changed blocks are copied back while the snapshot volume is being promoted to its parent disk. When Express Protect data is backed up to an HPE StoreOnce Appliance, it will be restored with the same number of streams used during the backup. Restore performance is dependent on the infrastructure.

HPE SimpliVity SimpliVity data protection technology differs significantly from conventional backup or snapshot methods. In many respects, it offers the best of both worlds: the complete protection of backup software with the speed and RPO/RTO of snapshots. Highlights include: No practical limit to the number of backups created and retained. HPE SimpliVity engineers have tested thousands of backups, and the upper tested maximum currently exceeds 200,000. Each copy is a full backup. There is no chain of changes that can be broken, corrupted, or needed for replay. Backups include a clone of the VM parent object. The full VM tree is logically copied in its entirety, and all VM trees are the same length. Size improves over time (that is, reduced) because all data, metadata, and pointers are deduplicated: Rehydration is not necessary as long as the data is on the HPE SimpliVity 380. Backup images can be transferred to an off-site system for DR. Two copies are automatically kept on different nodes. Copied blocks cannot be deleted. Data and metadata corruption can be detected with a "fingerprint" of checksum and hash values. Each node is functionally an independent system. Corruption on one node cannot affect another. If the original VM is deleted, it can still be recovered from the backup set. The HPE SimpliVity approach to data protection is to provide all the benefits of a full backup and data replication at the speed of a snapshot. The methodology is also an excellent defense against ransomware because all prior images are saved as full images, permitting recovery at a point before the ransomware (or any other) infection.

||||||||||||||||||||

||||||||||||||||||||

Because the HPE SimpliVity 380 deduplicates, compresses, and optimizes data the first time it is written to disk and maintains it in that state for its life cycle, backup copies are already deduplicated and compressed, thereby eliminating the need for third-party software or purpose-built backup appliances.

Preparing the backup design Optimizing the time required for the backup of a number of client systems and large databases that are all connected on different networks and platforms can be challenging. In addition, in business-critical environments, it is important to minimize the time needed for data recovery in case of a corrupt database or a disk disaster. The most common strategies for data protection, shown in Figure 10-15, include: Backups made to disk on-site, automatically copied to off-site disk, and finally copied to tape. Replication of data to an off-site location, which avoids the need to restore the data (only the systems need to be restored or synchronized). High-availability systems that keep both the data and system replicated off-site, enabling continuous access to systems and data. Backups made to tape and sent off-site at regular intervals. Differential backup which copies all files which have been modified or created since the last full backup, but does not modify the archive bit. Incremental backup which copies all files which have been modified or created since the last full or incremental backup and turns the archive bit off.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-15 A typical backup design Optimizing the time required for the backup of a number of client systems and large databases that are all connected on different networks and platforms can be challenging. In addition, in business-critical environments, it is important to minimize the time needed for data recovery in case of a corrupt database or a disk disaster. Be aware that whatever method you use, this must comply with the agreed RPO and RTO.

The infrastructure methods and devices BURA solutions combine the functionality and management of SANs, data protection and archiving software, and scaling tools to integrate tape and disk storage subsystems in the same SAN environment. Enterprise data protection can be accomplished with different target devices in various configurations using a variety of transport methods such as the corporate communication network, a server SCSI/SAS, SCSI, Fibre Channel over Ethernet (FCoE), or a Fibre Channel infrastructure. BURA solutions typically use a SAN that provides dedicated bandwidth independent of the LAN. This independence allows single or multiple backup or restore jobs to run without the network traffic caused by data protection environments. Management of the data protection and archiving software occurs over the LAN, and the data is sent over the SAN simultaneously. This achieves faster data transfer speeds and reduces Ethernet

||||||||||||||||||||

||||||||||||||||||||

traffic. Jobs and devices can be managed and viewed from either the primary server or any server or client connected within BURA solutions that have supported data protection and archiving software installed. All servers within the BURA solutions server group can display the same devices. Using hardware in parallel

Using several data paths in parallel is a fundamental and efficient method to improve performance. Parallelism boosts performance in these situations: Several client systems can be backed up locally, using the disks and related devices connected on the same client system. Several client systems can be backed up over the network. In this situation, the network traffic routing should be designed so that data paths do not overlap. Otherwise, the performance declines. Several objects (disks) can be backed up to one or several (tape) devices. Several dedicated network links between certain client systems can be used. For example, if system_A has six objects (disks) to be backed up and system_B has three fast tape devices, consider using three dedicated network links between system_A and system_B.

HPE data availability, protection, and retention compatibility matrix The HPE Data Availability, Protection, and Retention Compatibility Matrix provides information for designing data protection solutions, including backup/restore/archiving using storage products. The StoreOnce and StoreEver products have interoperability (IOP) details shown in their respective operating systems pages. The StoreOnce IOP pages show StoreOnce Virtual Tape Library support with Fibre Channel (FC), whereas the StoreEver IOP pages show both FC and SAS connectivity to tape products. Determine which HPE Products you are interested in finding support about, and then review that entire section for compatibility with HPE firmware, advanced features, connectivity compatibility, and Independent Software Vendors (ISVs). Consult the ISV product compatibility guide(s) for support details, such as product versions, OS versions, hotfixes, patches, special notes, driver details, and hardware devices supported by the vendor. See the ISV Compatibility Links page in this matrix for direct links to the vendor websites. Check the “Legacy Support” page (near the end of the matrix) to see if the product/OS

Technet24

||||||||||||||||||||

||||||||||||||||||||

has been moved to legacy support. If so, note the month at which support was moved to legacy support and find the older matrix link. If the matrix is not found online (because it is older than two years), email: "[email protected]" and request a copy of the older matrix. Because of differences in support, it is very important to verify which servers and OSs are supported by your HBA or controller by referring their individual QuickSpec. There are direct hyperlinks to HPE Server QuickSpecs in the StoreEver section in this Matrix. HPE QuickSpecs online: hpe.com/info/qs. HPE storage products should always run latest firmware, OS, drivers, and software versions. Systems running older versions face an increased risk from exposure to known (and unknown) issues. If a system is running older versions and encounters an issue, HPE Support will request upgrading to current versions. Each section of this matrix provides the current firmware, OS, drivers, or system software code that is supported by each platform. Additional technical information, guidelines, and rules can be found in the “HPE Backup, Recovery and Archive (BURA) Solutions design guide” and various Backup Application Implementation Guides and whitepapers located on the HPE Solution Whitepapers page in this matrix. For StoreEver and StoreOnce VTL backup, this Matrix only shows compatibility for FC SAN media servers. An FC SAN media server is defined as a server with direct access to backup devices. Other operating systems might be supported within Matrix by the respective ISVs as LAN clients, where a LAN client does not have direct access to the backup device but is backed up over a LAN connection to a server with direct access to backup devices. Receive support alerts (such as Customer Advisories), driver updates, software, firmware, and customer replaceable components, in your email through HPE Subscriber's Choice. Sign up for Subscriber's Choice Driver, Patch, Security, and Support alerts at: http://hpe.com/info/myadvisory For usage of non-HPE branded HBAs and switches, HPE recommends that such hardware be made “HPE-like”, through the deployment of same versions of firmware, BIOS, NVRAM, configurations, and, where applicable, HBA drivers. Note Read more in the HPE Data Availability, Protection and Retention Compatibility matrix at https://support.hpe.com/hpsc/doc/public/display?docId=c04616269

||||||||||||||||||||

Capacity and sizing The required storage capacity for backups depends on many factors including the following: The amount of backup data written and retained—size and number of protected clients –

How many backups are retained (recovery points needed)

–

Type of backups (full, incremental, and differential)

How much the data is compacted (deduplicated and compressed)—the type of data –

Rate of change between backups

–

Configuration options such as block size, multistreaming, and multiplexing

–

The efficiency of deduplication and compression

||||||||||||||||||||

Data compaction refers to the removal of redundant information from a backup set before storing on a backup device. Incremental backups, deduplication, and compression are all methods for removing redundant data from a backup set. The deduplication ratio indicates how much new information is backed up to the HPE StoreOnce appliance versus the size of the actual data coming in. After numerous sequential backups, the amount of data stored on disk can be reduced. Generally, after multiple backups, the deduplication ratio starts to increase rapidly, which indicates less and less information has to be backed up and stored on disk. An administrator might desire two weeks of backups stored on an HPE StoreOnce System for quick recovery access. Data deduplication provides more backup space without increasing the physical capacity of the backup device; however, a dynamic environment with changing data affects the backup data deduplication ratio. In performing these tests, a combination of servers were used, running various operating systems and using various data types that contained customer representative data sets with realistic structure and content. Data was updated between each backup until the desired rate of change was reached. Weekly full and daily incremental backups

Many backup environments use a weekly full and daily incremental backup schedule. Some characteristics of full and incremental backups are: Full backups include both the changed and unchanged data in a data set. Incremental backups include only changed data in a data set.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Changed data does not deduplicate as much as unchanged data, so incremental backups do not deduplicate as much as full backups. Incremental backups are usually much quicker than full backups and use fewer resources, which results in less impact to the backup server, network, disk storage device, and HPE StoreOnce System. End-to-end data compaction is greater for schedules that include incremental backups when compared with daily full backup schedules. The result is greater space saving on the StoreOnce Backup appliance even though deduplication ratios, measured on the StoreOnce Backup system, are lower for incremental backups. Full backups enable faster and simpler recovery than incremental backups or a mix of full and incremental backups.

Capacity planning usage models The first column in Figure 10-16 indicates the size of the backup, and the second is the space saved when used on the StoreOnce.

Figure 10-16 Capacity planning usage models An environment with two servers with 300 GB of data each and a 14-day backup data

||||||||||||||||||||

retention requirement can have several StoreOnce Backup appliance usage models. Usage models change based on parameters such as the following: Backup schedule type: –

Daily full image backups deduplicate well but use more backup infrastructure and StoreOnce compute and bandwidth resources during a backup.

–

Weekly full with daily incremental backups do not deduplicate as well but use less compute and bandwidth resources.

–

The end-to-end data compaction for weekly full with daily incremental backup schedules might be better than daily full backups.

||||||||||||||||||||

Sequential or simultaneous backups

Multiple backups running simultaneously typically have better backup throughput but might reduce StoreOnce deduplication ratios. In Figure 10-16, we compare five usage models for backups to a StoreOnce System with the following common characteristics: Data type: flat file Data size: 300 GB Backup schedule: daily Retention period: 14 days Each usage model shows the overall size of the backup data without deduplication vs. the size of the data on the StoreOnce System after deduplication.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check What three steps help in defining a backup strategy?

||||||||||||||||||||

||||||||||||||||||||

Array snapshot with HPE RMC-V Impact of traditional backup on mission-critical applications in all-flash 3PAR environments Figure 10-17 illustrates four points that can impact your production environment in a backup: when you put the application into backup mode when you start to transfer the data when the data transfer is complete when you take the app out of backup mode

Figure 10-17 Traditional backup (full) With a traditional full backup, the application is impacted for the full time that the data is transferred—and with a full backup of a large database that can be quite a long backup window.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Impact of an incremental backup on mission-critical applications in all-flash 3PAR environments For this reason, most full backups are done at the weekend with incremental backup, shown in Figure 10-18, deployed during the week—with an incremental backup, you tighten the backup window because the time that data is transferred is much shorter— however, recovery from an incremental backup can be a lot more complex and time consuming.

Figure 10-18 Traditional backup (incremental)

Eliminating backup impact on mission-critical applications in all-flash 3PAR environments with 3PAR RMC 3PAR RMC looks completely different from traditional full backup and incrementals— the only impact on the application server takes place when the application is quiesced momentarily while a consistent, nondisruptive snapshot is read on 3PAR—and for the entire time the data is transferred, there is no impact on the application—you get a full backup in the fraction of the time and not have any impact on the application server, as shown in Figure 10-19.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-19 3PAR RMC

Snapshots alone cannot deliver comprehensive application protection in all-flash environments One way that IT organizations have (at least partially) addressed the limitations of the traditional server-based backup is to use array-based snapshots—in conjunction with replication—to meet stringent RPO and RTO requirements. In high-availability virtual-flash environments, snapshots are typically the first line of defense against data loss. Flash array-based snapshots and replication provide fast, nondisruptive point-in-time copies of data. However, as shown in Figure 10-20, snapshots alone cannot deliver comprehensive backup as they have retention limitations, corruption vulnerabilities, and (relying 100% on the original data) a dependence on the underlying flash storage system. Simply put, snapshots will be lost if the flash storage system fails. Snapshots are thus designed to complement as opposed to replace backup.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-20 Snapshots are not backups A more effective approach to protecting data on flash arrays is to combine the near instant, nonintrusive availability of local and remote snapshots with the reliable recovery and cost-effective retention of backups, delivered in an application-aware, flash-integrated data protection solution.

Protecting workloads at the speed of 3PAR Flash As flash expands across the data center, protecting and securing data at speeds not possible with traditional methods has become a critical requirement. Backups need to be near instantaneous, frequent, nondisruptive, and application consistent. 3PAR Express Protect delivers the performance that 3PAR flash application SLAs require with 23 times faster backups and five times faster restores than traditional backup methods. It achieves this by delivering a 3PAR flash-integrated backup that protects data by copying it directly from its source (3PAR) to protection storage (StoreOnce) via the most efficient path without application impact and independent of backup server software. Optimized data movement

When a backup is triggered, Express Protect creates a snapshot on the 3PAR array, as

||||||||||||||||||||

||||||||||||||||||||

shown in Figure 10-21. This is a simple process and decoupled from the actual movement of data, so the application only needs to be in backup mode for a brief instant. 3PAR Express Protect then moves the snapshot data—using the HPE StoreOnce Catalyst Protocol—to the StoreOnce Appliance. Unlike a traditional backup application, 3PAR Express Protect has full access to the 3PAR SnapDiff block-change detection technology. It therefore knows exactly what has changed since the last backup and only has to send those unique changed blocks at the array level across the network to the StoreOnce Appliance. 3PAR Express Protect also ensures “low-bandwidth transfer” by subdividing these blocks into 4 KB chunks and checking with the StoreOnce Appliance to see which blocks are new and need to be moved to the store. During this process, no host resources are used, virtually eliminating performance impact on the application.

Figure 10-21 Deliver on protection SLAs with no compromise flash performance Full recovery at the cost of an incremental

3PAR Express Protect differentiates from traditional backup solutions in the way the data is processed and stored by StoreOnce. When StoreOnce receives the changed blocks from 3PAR, the StoreOnce indexing process uses pointers to map previously written blocks along with the new ones so that a synthetic full copy of the data is always available for instant recovery whenever a restore is required. This is in direct contrast to certain traditional backup methods that expect you to recover a full backup and subsequently recover a series of incremental backups to restore the application data back to a certain point in time. Thus, 3PAR Express Protect provides the performance of

Technet24

||||||||||||||||||||

||||||||||||||||||||

incremental backups for what is essentially equivalent to a full back up as well as ensuring a fast and simple recovery that saves the user a tremendous amount of time. Multistreaming support

Another key element of 3PAR Express Protect is the ability to split backups to StoreOnce into multiple streams to improve performance. By breaking up the application data into separate streams, 3PAR Express Protect can write concurrently to different objects within a StoreOnce Catalyst Store, which helps combat the performance issues typically found with full backups. Multistreaming also helps with the delta-change “incremental” in Express Protect synthetic full, in scenarios when high change rates might mean significant numbers of changed blocks would be backed up.

Flash-integrated data protection without compromise Flash-integrated data protection means that primary flash storage and backup appliances can be integrated with a simple software management solution, common data services, and automation between devices enabling seamless data movement. Data protection becomes a function of primary storage, eliminating the need for additional backup infrastructures (media servers). This makes protecting your data less intrusive on application processing, simpler to manage, and faster to complete. Removing the layers of complexity leaves you with a simple backup process that can provide fully automated protection of your primary storage arrays managed directly from your hypervisor or application interface. Data moves natively from primary storage to backup as scheduled by the business application owner, without the need for media servers or complex backup software. If flash is the future of storage, then the future of data protection is flash-integrated data protection, as shown in Figure 10-22.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-22 Flash-integrated data protection without compromise

Sizing an RMC solution RMC performance, sizing, and placement considerations: Determine the number of streams that need to run in parallel. Host the RMC VM on a host that can support that level of performance. Ensure that the host can support the required transport (to array(s) and StoreOnce devices). Match the StoreOnce to the streams and capacity required. RMC does not use any special features found only in multinode systems. What host should the RMC VM run on. Take into consideration volume sizes, change rates, retention required, and performance requirements A single RMC VM can handle a maximum 256 simultaneous streams. Consider a 1:1 relationship between vCPU and physical cores for maximum performance. Each (maximum config) RMC appliance uses 64GB. This memory has to be reserved in the case of PCI Passthrough.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Leave some overhead for the Hypervisor. Formula for calculation of number of RMC instances needed:n = (v * s)/Smax where: n = Number of RMC required v = Number of 3PAR Virtual Volumes s = Stream count per volume Smax = Maximum streams per RMC appliance

Streams per RMC Appliance Small configuration are designed for snapshot management only, so no Express Protect or Peer Copy is possible, as shown in Figure 10-23.

Figure 10-23 Streams per RMC Appliance

Sizing at deployment time For each selected “CPU and Memory Configuration,” the wizard will tell you what such a configuration will allow you, as shown in Figure 10-24, Figure 10-25, Figure 10-26, and Figure 10-27.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-24 Example of selected configuration in RMC Installer (small configuration)

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-25 Example of selected configuration in RMC Installer (medium configuration)

Figure 10-26 Example of selected configuration in RMC Installer (large configuration)

Figure 10-27 Example of selected configuration in RMC Installer (extra large configuration)

Max streams per StoreOnce (theoretical) The amount of concurrent data streams a StoreOnce Appliance can operate at, as shown in Figure 10-28, does not necessarily increase performance in a linear or exponential manner.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-28 Theoretical maximum number of streams per StoreOnce backup system Where performance is concerned, an increase in the amount of streams per volume will show benefits up until a limit, but the overall StoreOnce Appliance performance will not exceed a certain amount. Increasing the number of shelves (5100 and above) increases the throughput in terms of speed, but the figure for maximum concurrent streams remains the same.

Capacity sizing Capacity sizing is shown in Figure 10-29 (StoreOnce 3.15 was used).

Figure 10-29 Capacity Sizing

Sizing the RMC host RMC host sizing and recommendations, shown in Figure 10-30:

Technet24

||||||||||||||||||||

||||||||||||||||||||

Determine the number of RMC instances that are needed (n)—calculated earlier n*16 = The number of physical cores consumed by RMC n*64 = The amount of RAM reserved by RMC n = The amount of PCI slots to make available to RMC for FC passthrough Take into account the overhead for running the hypervisor (ESXi detailed below)

Figure 10-30 RMC host sizing and recommendations Sizing will differ on Hyper-V.

Sizing recommendations—ProLiant servers In the higher specification machines, it is the number of PCI slots and therefore FC ports available that is the limiting factor for the number of RMC instances that can be run (if you are making use of FC). If customer has Blade infrastructure running applications, it is possible to use a server from Figure 10-31 to host RMC, separate from the blades, if the configuration of the blades is not suitable for RMC hosting.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-31 Sizing recommendations—ProLiant servers

Technet24

||||||||||||||||||||

Learning check When sizing the RMC host, how many vCPUs do you need to support Peer Copy and Express Protect? A. 1 B. 2 C. 3 D. 4

||||||||||||||||||||

||||||||||||||||||||

||||||||||||||||||||

Software to support HPE BURA HPE software for BURA includes HPE Store Once Catalyst, Data Protector, RMC, and Scality for an end-to-end data protection solution. However, in many cases, customers prefer a third-party ISV solution besides Data Protector, as shown in Figure 10-32. HPE collaborates with several key market-leading software vendors to deliver a data protection solution integrated with HPE BURA products. Leveraging extensive partnerships with leading software companies, HPE continues to provide software solutions that support the backup and restore processes of homogeneous and heterogeneous operating systems in a shared storage environment.

Figure 10-32 Software to support HPE BURA

Technet24

||||||||||||||||||||

||||||||||||||||||||

It is important to understand how HPE BURA solutions integrate with these third-party solutions. Traditional ISV: Micro Focus Data Protector Veritas NetBackup Veritas Backup Exec Veeam CommVault Data Platform (formerly Simpana) IBM TSM (also known as Spectrum Protect) EMC Networker Application direct backup Oracle RMAN MS SQL SAP HANA

Traditional backup and archive is unsustainable Many storage vendors continue to design and produce storage architectures that have done an excellent job in storing, protecting, and retaining data up until now. However, these designs are starting to “creak,” and they might not support the changes that are going to happen in IT over the next 5 to 10 years. If we consider the architectures on Figure 10-33, they are siloed, limited in terms of their ability to scale both capacity and performance; when customers outgrow them, it will be a “forklift upgrade” with a brand new and very costly solution.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-33 Solutions must be future ready The same is true of the backup architectures. They are good solutions to date, but they are based on what we at HPE sometimes refer to as “Deduplication 1.0” architectures. They are siloed, homogeneous, deduplication engines, which are focused on specific and very limiting use cases—one use case for data centers and another use case for remote offices. Not only are current backup strategy solutions fragmented and complex, but when we think about the notion of backup and recovery solutions, they are also built for specific traditional infrastructures. However, the challenge is that the IT organization has to be

Technet24

||||||||||||||||||||

||||||||||||||||||||

dynamic and agile enough to not only grow with the business but also to allow the business to respond to the market. If we consider how data will grow over the next few years, there is enormous growth with 50% of that data typically residing outside of the traditional boundaries of the main data center. To give context to that 50x growth, it equates to 40 zettabytes of data existing in the world; yet, if we analyze this, we find that around only 33% of the data is something that can be distilled into usable and valuable information. The rest is normal communications between people or infrastructure components. Therefore, why would someone make that 67% (which equates to just over 26 zettabytes) part of the backup strategy? The answer is that the person should not do so, and neither should our customers. This is because the bottom line is that for organizations to succeed, they have to focus on how they will standardize all aspects of their data center. Also, just as importantly, they must generalize their resources. By generalize, we mean that by using fewer specialized staff, there will be a smaller premium needed to pay for them. Another issue driving the changes to a more flexible architecture is that legacy systems cannot scale, even to the point of getting to 33% of the market. However, by addressing a converged approach, we can help organizations scale much more easily. As part of standardization, what typically was traditional data protection might have been an afterthought. However, it no longer can be an afterthought. Now, it is equally as important to protect data residing on primary storage as well as in regional and branch offices. We also start to find many primary applications, data bases, and so on with backup integrated into them.

HPE BURA

||||||||||||||||||||

||||||||||||||||||||

Figure 10-34 Backup recovery and archive Figure 10-34 might look a little complex, but it is a fairly typical deployment model for many organizations. The primary data center is connected to the regional offices and to the smaller branch offices (which might be sales offices, for example). The primary data center is also connected to one or more disaster recovery sites. These solutions are not bound to just the primary data center. They cannot be. Typically, 50% of an organization’s data resides outside that data center. For this reason, an ideal solution must encompass any place that the data is generated and whether the organization uses and relies on the data anywhere from the regional offices to the smallest remote site. All organizations evolve and often merge. When customers try to integrate a new company, it brings new systems and new processes. For example, a company might have 14 different backup processes and solutions because, over time, they had grown and merged but did not consolidate their data protection processes. The environment might not be that complex, but if customers apply the BURA perspective to the way they approach the data protection strategy, regardless of different business opportunities and processes, the speed at which customers can protect and Technet24

||||||||||||||||||||

||||||||||||||||||||

restore that data can be assured. The value customers can extract from it could be immense, while at the same time, driving down cost and reducing risk. The data on the 3PAR infrastructure could be moved to archiving devices, HPE’s scaleout archiving platform, and Autonomy Consolidated Archive, to provide lower-cost and longer-term retention of data. It would also provide incredibly fast extraction of value from that data for business intelligence purposes. StoreOnce deduplication solutions can be deployed with Data Protector from the very smallest remote sites (containing just application servers, virtualized or nonvirtualized) to the largest data center or disaster recovery site. This would be one single completely compatible backup solution. All of this data can be offloaded to tape for very low-cost disaster recovery, compliance, and archival purposes. All of this can be configured and managed from central data center, reducing branch office staffing costs, reducing space, reducing network traffic, and, more importantly, reducing costs and risk. At the same time, it provides fast data recovery in the event of system failure and extracts maximum value from the data for marketing or business intelligence purposes. Before the ideal approach just described can be achieved, another topic must be addressed: the steps to classify the information that an organization generates. This will help result in a successful data protection and retention strategy.

Data Protector Micro Focus Data Protector, shown in Figure 10-35, delivers comprehensive data protection, real-time intelligence, and guided optimization to ensure reliable and costeffective backup and recovery. This solution standardizes the backup and recovery of information spread across locations, applications, formats, storage platforms, operating systems, and hypervisors.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-35 Data Protector Built on a unified architecture that leverages analytics and automation, Micro Focus Data Protector provides assurance for mission-critical information from the core to the edge, across physical, virtual, and cloud infrastructures. When combined with its companion products in the HPE Adaptive Backup and Recovery Suite, Data Protector becomes a solution that allows to adapt and optimize your environment at all phases of the data protection cycle to meet your operational and business goals.

Data Protector 10.0 A next-generation GUI will take over from the previous GUI in a number of steps. The previous GUI is still needed to Create backup/copy specs, media/device management, and deployment/updates. The new GUI offers a Dashboard, Telemetry setup, and info, as well as a new consolidated scheduler. Micro Focus Data Protector 10.0, shown in Figure 10-36, provides Security Enhancements, new user interface, web-based scheduler support, REST API management, and New Platform support Security enhancements—all communication are secured between cell managers and clients. Push installation of clients is using SSH protocol for Linux and UNIX clients. All commands must now be executed by the Cell manager ensuring both control and data are secured via the TLS 1.2 channel and guarantee data integrity.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-36 Data Protector 10.0

Data Protector Architecture Key components of the architecture of Data Protector, as shown in Figure 10-37, include: Client systems—Clients to be backed up must have the Data Protector disk agent, also called the backup agent, installed. To back up online database integrations, install the application agent. The disk agent reads or writes data from a disk on the system and sends or receives data from a media agent. The disk agent is also installed on the Cell Manager, allowing customers to back up data from the Cell Manager, including the Data Protector configuration, and the internal database (IDB). Note The term disk agent is used for both agents.

Systems with backup devices—Clients with connected backup devices must have a Data Protector media agent installed. Such systems are called backup or media servers. You can connect a backup device to any system, not only to the Cell Manager. The media agent reads and writes data to and from media in the device and sends or receives data from the disk agent. Systems with a user interface—You can manage Data Protector from any system on the network on which the Data Protector GUI and Command Line Interface (CLI) are installed. As a result, the Cell Manager system can run in a data center and Data

||||||||||||||||||||

||||||||||||||||||||

Protector can be managed from a desktop system. Installation Server—The Installation Server holds a repository of the Data Protector software packages for a specific architecture. By default, the Cell Manager is also an Installation Server. At least two Installation Servers are needed for a mixed environment, one for UNIX and one for Windows. Internal database—The IDB, which is located on the Cell Manager, holds information about what data is backed up, which media the data resides on, and which devices and libraries are configured. The database also stores the results of backup, restore, object copy, object consolidation, object verification, and media management sessions.

Figure 10-37 Data Protector architecture

How deduplication integrates into Data Protector The StoreOnce deduplication engine is located on Data Protector clients or dedicated gateway servers, as shown in Figure 10-38. The deduplication engine writes deduplicated data (low bandwidth) to a StoreOnce library or a StoreOnce backup system.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-38 How deduplication integrates into Data Protector A deduplication client library is on the media agent client as part of the media agent. A gateway client has a software deduplication (HPE StoreOnce Catalyst software) media agent component installed. In general terms, data deduplication can operate at the file, block, and even at the bit level. File deduplication eliminates duplicate files but is not a very efficient method of deduplication. Block and bit deduplication looks within a file and saves unique iterations of each block or bit. Each chunk of data is processed using a hash algorithm such as MD5 or SHA-1. With data deduplication, only one instance of the data is actually stored; each subsequent instance is just referenced back to the one saved copy. For example, a 100MB storage demand could be reduced to only 1MB. The deduplication ratio depends on the type of data and affinity the data has with the new data. Data deduplication offers other benefits. Lower storage requirements reflect on disk or tape expenses. Deduplication optimizes disk space and allows for longer disk retention

||||||||||||||||||||

||||||||||||||||||||

periods. Longer retention periods allow for better recovery time objectives, greatly reducing the need for tape backups. Finally, data deduplication reduces the data that is sent over the slow network (such as WAN) and greatly improves remote backup, replication, and disaster recovery capabilities.

Target-side deduplication with Data Protector and StoreOnce The deduplication process takes place on the target device. The target device receives data to be backed up from media agents installed on clients (gateways). The StoreOnce software deduplication system then writes the duplicated data to the StoreOnce library. (This is the physical store and is sometimes referred to as the deduplication store.) The StoreOnce software deduplication system allows connections from several media agents, locally or remotely, as shown in Figure 10-39. It also provides synchronization mechanisms to enable multiple media agents to work with the StoreOnce library at the same time.

Figure 10-39 Target-side deduplication with Data Protector and StoreOnce

Server-side deduplication with Data Protector and Technet24

||||||||||||||||||||

||||||||||||||||||||

StoreOnce Server-side deduplication enables the duplication of data from clients on which deduplication is not supported locally. The media agent performs server-side deduplication on a separate media agent client (called a gateway), as shown in Figure 10-40. This reduces the load on the backed up system and on the target device, but it does not reduce the amount of network traffic between the disk agent and media agent. Note The media agent client must support deduplication: 64-bit Windows systems or 64-bit Linux systems only. To access the Data Agile BURA Compatibility Matrix, go to: https://support.hpe.com/hpsc/doc/public/display?docId=c04616269

Figure 10-40 Server-side deduplication with Data Protector and StoreOnce The decision whether to use source or server deduplication depends on the WAN or LAN link speed available, CPU and memory size of the server running the disk agent, and how much load that server can withstand. If, for example, a customer has a large database server that is heavily loaded and they are backing up a snapshot, it might not be a good idea to use source- or server-side deduplication for some jobs. Target-side deduplication places fewer loads on the server.

||||||||||||||||||||

||||||||||||||||||||

Small remote office backups over a WAN might benefit from server-side deduplication because their data change rate is low and they can be backed up outside of peak hours when server load is not so critical.

Source-side deduplication with Data Protector and StoreOnce With source-side deduplication, a media agent is installed together with the disk agent on the client that is backed up. Thus, the client becomes a source-side gateway, as shown in Figure 10-41. The media agent performs deduplication on the client itself, so only duplicated data is sent to the target device, thereby reducing overall network traffic. The number of concurrent streams is limited by load balancing settings. After a media agent finishes the backup of local objects, a new media agent is started on the next client system.

Figure 10-41 Source-side deduplication with Data Protector and StoreOnce The media agent reads or writes data in terms of object versions to or from the StoreOnce library. Each object version is represented as an item in the StoreOnce library. To optimize deduplication performance, disk agent concurrency is not supported. (This means one disk agent talks to one media agent—there is no multiplexing of streams.)

Technet24

||||||||||||||||||||

Note From the Data Protector perspective, the deduplication process looks very similar to target-side deduplication using the StoreOnce software system. However, there is no separate StoreOnce software deduplication system, and the deduplication takes place on the StoreOnce Backup system device itself.

Capacity planning considerations Information needed per node To continue with your capacity planning, you would need to collect this type of information: Bus/adapter(s): –

Maximum bandwidth (MB/s)

–

Maximum throughput (IO/s)

–

Available bandwidth

–

Available throughput

Storage (per logical drive): –

Total capacity (in GB)

–

Used capacity (for full backups)

–

% daily change (for incremental backups)

–

% overlap (for differential backups)

–

Type of data (for compression calculations)

–

Maximum data transfer rate

–

Maximum throughput

–

Available bandwidth

–

Available throughput

Network connection: –

||||||||||||||||||||

Maximum bandwidth

||||||||||||||||||||

–

Available bandwidth

–

% protocol overhead

Maximum is used to indicate the upper limit of a given device, not the theoretical limits of the technology used (for example, FC-AL has a theoretical limit of approximately 50,000 IO/s; however, a given FC-AL adapter might be capable of less than 50% of this rate). Available is used to indicate the resource capability of a given device not currently in use. Is there any other type of information that you believe you need to collect? If so, indicate it below:

Information needed per network device Information you need to collect for network hubs and switches includes: Maximum transfer rate Maximum throughput Available bandwidth Available throughput Number of ports; for each port: –

Maximum transfer rate

–

Maximum throughput

–

Available bandwidth

–

Available throughput

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Is there any other type of information that you believe you need to collect? If so, indicate it below:

Information needed per backup server For each backup job, you need to collect: Which clients are being backed up? Which files are being backed up? What is the data compression rate? Which rotation scheme should be used? What is the frequency, protection period, and append rules? Which backup type (full, differential, or incremental) should be used and when? Miscellaneous questions might include: What is the expected data growth rate? What is the data change rate? Is there any other type of information that you believe you need to collect? If so, indicate it below:

||||||||||||||||||||

Capacity planning calculations As part of the capacity planning calculations, you should determine: The data arrival rate at the backup server(s) The number of tape drives needed The number of tape media needed The duration of the backup window At the same time, you might be able to identify possible bottlenecks in the device chain.

Activity: Sample backup information collection for capacity planning Instructions: Read the scenario and answer the question. Scenario: Your customer asked you to prepare a proposal for a backup solution. You are analyzing the backup performance requirements and planning for the right capacity. You wrote these questions in preparation for your meeting with the customer: –

How long will it take to back up the customer environment? In other words, what is the backup window?

–

How much data must be backed up?

–

How many tapes are needed?

–

How many tape drives are necessary?

||||||||||||||||||||

Answer the following question: What other questions should you ask?

Technet24

||||||||||||||||||||

||||||||||||||||||||

Micro Focus Adaptive Backup and Recovery Micro Focus will continue to support products from HPE Software, including: Adaptive Backup and Recovery suite This is composed of one core and two companion products that leverage insight throughout the continuum of the data protection cycle—optimize, protect, and analyze through information insight, as shown in Figure 10-42. It allows for a 360 degree analysis of the backup environment and of the end-to-end protection process via insights, enabling users to adapt and tune the process so that it can run optimally at all time. Storage Optimizer—Analyze and manage data based on its value. Reduce costs, meet backup SLAs, and reduce compliance risks. Data protector—Centralized and standardized protection across repositories, platforms, and locations. Advanced backup and recovery options, compression, and deduplication. Backup Navigator—Advanced analytics, reporting, and monitoring. Find protection gaps, root-cause analysis, backup inefficiencies, remediation, and capacity planning.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-42 Micro Focus Adaptive Backup and Recovery

HPE Backup Navigator HPE Backup Navigator, shown in Figure 10-43, is a highly interactive and intuitive reporting application that combines strong visualization and real-time operational analytics to simplify and optimize the backup and recovery experience for HPE Data Protector environments.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-43 HPE Backup Navigator Administrators and auditors can monitor Data Protector environments and generate reports using the Backup Navigator dashboard. This product is very scalable and can be used by customers ranging from small- and medium-sized businesses to large enterprises. What is new in HPE Backup Navigator 9.60

Support for the latest version of Data Protector version 10.0. Integration with VM Explorer allowing management of both Data Protector and VM Explorer through one reporting tool. Performance improvements for data collection and reporting, enhanced monitoring for performance This release adds detailed automatic problem detection and recommended actions to identify and resolve issues quicker by intelligent reporting. Support for reporting scheduled backup sessions created using the new web-based scheduler available with Data Protector 10.0. Flexibility to edit and modify all custom reports created using the Custom Report Wizard.

||||||||||||||||||||

||||||||||||||||||||

Backup Navigator uses analytics to improve backup and recovery processes. It provides: Real-time predictive analytics—Customers can gain visual foresight into backup and recovery processes based on daily use. Trending and forecasting algorithms reveal future performance and capacity gaps and requirements specific to dataset characteristics, infrastructure capabilities, and organizational requirements. Rapid root-cause analysis and problem solving—This solution proactively detects and addresses potential resource conflicts and systematic or systemic issues before they cascade into outages or data loss that negatively impacts business operations. Collaboration and cross-system support—An automated report creation process schedules and shares reports with organizational stakeholders. Reports can also be securely isolated and made available to external customers. The same information can be exported in a variety of formats for inclusion into other organizational systems. Extensible reports can be customized and tailored to match the specific needs of the operator, organization, or customer. “What-if” scenario evaluation—Get intelligent insights into current backup and recovery operations before new data sets are included in the process. Such insights can identify whether or not SLAs are achievable, the impacts to the backup infrastructure (for example, physical capacity, network load, and device loads), and the best ways to balance the demands of new data sets within the existing infrastructure.

Recovery Manager Central and RMC-V RMC-V is the HPE Storage product that leverages HPE 3PAR/StoreVirtual VSA array snapshots and HPE StoreOnce backup appliances to implement fast and efficient Data Protection. Veeam Explorers are tools that extend the functionality of Veeam Backup & Replication by allowing you to view and recover application items from within the application VM backup or replica. RMC-V ERT is the data recovery technology that can export the array snapshot and/or backup images stored in StoreOnce (called Express Protect) as a disk format for filesystem item recovery. By leveraging Veeam Explorers for Active Directory (AD)/MSSQL/SharePoint/Oracle/Exchange, it is possible to recover application items (granular) directly from the RMC-V Express Protect backup.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Recovery Manager Central and RMC-V are shown in Figure 10-44.

Figure 10-44 Recovery Manager Central and RMC-V

Veeam and StoreOnce integration Veeam Backup and Replication integrates with StoreOnce Catalyst, as shown in Figure 10-45, to enable source-side deduplication, virtual synthetic full backups, and faster restores. This leads to more efficient backup data transfer over the network, reduced storage space, and faster performance in comparison to backup targets with traditional NAS protocol (CIFS and NFS) connectivity.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-45 Veeam and StoreOnce integration

Veeam using Catalyst for source-side deduplication over Fibre Channel HPE StoreOnce can be used as a Veeam Backup & Replication target in Backup over LAN or LAN-free backup. Veeam can use Catalyst for source-side deduplication over fibre channel, as shown in Figure 10-46.

Figure 10-46 Veeam using Catalyst for source-side deduplication over Fibre Channel It is a good practice to restrict access to the Catalyst stores only to the clients that should write and read backup data from the store. StoreOnce Catalyst Client access permission checking is used to control this access and is implemented by creating Catalyst client identifiers. No Client Access Permission Checking allows all clients to have access to the StoreOnce Catalyst store.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Client Access Permission Checking allows a selected group of client identifiers to access to the StoreOnce Catalyst store. Software and security requirements:

Veeam Backup & Replication Enterprise or Enterprise Plus version 9 Update 2 or later with StoreOnce 3.13.1 or later for general Catalyst integration. Veeam Backup & Replication Enterprise or Enterprise Plus version 9.5 or later with StoreOnce 3.15.1 or later is required for Instant VM Recovery® and other vPower® NFS-based features. For full compatibility details, please see the support matrix at http://www.hpe.com/storage/DAPRCompatibility Recommended—10GbE, 8Gb FC, or 16Gb FC connectivity from the Veeam Gateway server to the StoreOnce appliance.

Where to download Veeam Backup Free Edition

Figure 10-47 Veeam® Backup Free Edition Veeam® Backup Free Edition is a must-have free tool for ad-hoc virtual machine (VM) backup, restore and management in VMware vSphere and Microsoft Hyper-V virtual environments. With no agents to deploy, powerful recovery options and VeeamZIP™, administrators get the flexibility in options and a reliable solution for important VMs.

||||||||||||||||||||

||||||||||||||||||||

Note You can quickly back up running and powered off VMs with VeeamZIP. Creating a VeeamZIP file can be helpful if you want create an ad-hoc backup for VMs, archive VMs before decommissioning, and so on. You can create a VeeamZIP file for one or more VMs. (March 2018, helpcenter.veeam.com) Whether you are locked in a legacy backup contract or are operating within a tight budget, you will find this free VM backup software helpful for your daily VM management. Veeam Backup Free Edition works perfectly with both standalone ESXi and Hyper-V hosts, as well as with large virtual deployments. Whether you run a test lab or a production environment, it will be a perfect match. Note More information about restore and management in VMware vSphere and Microsoft Hyper-V virtual environments can be found at: https://www.veeam.com/vmware-esx-backup.html https://hyperv.veeam.com/ Are you on a budget or still experimenting with VM backup? Veeam Backup Free Edition is the perfect solution because it is: Powerful: Clone, copy, export, and manage your VMware and Hyper-V VMs. Look inside the VM to restore individual application items Easy to use: No complicated configuration and no need to power off the VM—It Just Works!™ FREE: Unlimited number of ESXi and Hyper-V hosts or VMs, with no expiration date and zero investment required

Comparison of several item recovery options within Veeam in HPE Storage environment Veeam® Availability Suite™ delivers Availability for the Always-On Enterprise™ by combining the industry-leading backup, restore and replication capabilities of Veeam

Technet24

||||||||||||||||||||

||||||||||||||||||||

Backup & Replication and Veeam ONE™ for VMware vSphere and Microsoft Hyper-V. Veeam offers different editions of Veeam Availability Suite—Standard, Enterprise, and Enterprise Plus—to give you the functionality your business needs. Comparison of several item recovery options within Veeam in HPE Storage environment is shown in Figure 10-48. Note To access Veeam pricing and packaging information, go to: https://www.veeam.com/availability-suite-pricing.html

Figure 10-48 Comparison of several item recovery options within Veeam in HPE Storage environment

Integrate 3PAR Integration with Veeam To integrate HPE 3PAR StoreServ with Veeam, select “Storage Infrastructure” in lefthand pane of Veeam Backup and Replication console, select Storage Infrastructure, and click Add Storage, as shown in Figure 10-49.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-49 Veeam—Add 3PAR Integration

Veeam with Nimble highlights The cost of downtime and lost data poses a significant challenge to meeting SLAs for the Availability of apps and data. Veeam delivers Availability for the Always-On Enterprise through integrations with enterprise storage offerings, now including Nimble Storage in Veeam Availability Suite 9.5, as shown in Figure 10-50 and Figure Figure 10-51. Nimble with Veeam offers dramatically faster, more reliable data protection and management of array-based snapshots and replication through the Veeam GUI. Space-efficient protection is built into Nimble Unified Flash Fabric: snapshots, replication, and cloning.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-50 Veeam with Nimble highlights

Figure 10-51 Nimble Storage Solution for Veeam Availability Suite Reduce the impact on your production environment and achieve lower recovery time and point objectives (RTPO) with Veeam Backup from Storage Snapshots and Veeam Explorer for Storage Snapshots with all Nimble Storage: Backup from Storage Snapshots: Lower the additional impact from backup activities on

||||||||||||||||||||

||||||||||||||||||||

your production storage by retrieving VM data from Nimble Replicated Copy secondary storage systems, instead of from primary storage. Veeam Explorer for Nimble Storage Snapshots: Recover individual items or entire VMs quickly and efficiently from Nimble Snapshots and Replicated Copies. Take advantage of the low overhead that periodic SAN snapshots offer and perform granular recovery from virtual machines. Veeam Backup & Replication automates the entire process to recover individual files, items, or entire VMs in two minutes or less—without staging or intermediate steps. On-Demand Sandbox for Storage Snapshots: Use storage snapshots on primary and secondary storage systems to create complete isolated copies of your production environment in just a few clicks, for fast and easy developing, testing, and troubleshooting. Faster, Reliable Data Protection: Veeam Backup & Replication is integrated with Nimble to improve functionality and ease of use: Admins can now leverage and manage array-based snapshots on Veeam-protected Nimble arrays and control them from within the Veeam GUI. Data protection can be automated with VM-level granularity, and the recovery of individual files, items, or entire VMs can be done quickly—in two minutes or less—without staging or intermediate steps. In addition, Nimble Storage arrays can store thousands of snapshots with no performance penalty. Superior Data Efficiency: Nimble Storage has data reduction that can deliver 10x space savings or more. This allows you to deploy increased data protection that matches the needs of your business without worrying about the cost of ballooning capacity. Data reduction capabilities include content-aware dedupe on All Flash that delivers 5X more effective capacity, variable block compression, zero pattern elimination, and zero copy clones that can clone snapshots without any added capacity use.

Backup from Storage snapshots Taking VMware vSphere snapshots can produce a serious burden on virtual machine (VM) performance, and it can take considerable effort by administrators to overcome this technical challenge and meet the required SLAs. Veeam dramatically improves this process, providing built-in integration with production storage, as shown in Figure 1052, leveraging storage snapshot functionality to reduce the impact on the environment from vSphere snapshot removal during backup and replication. Don’t be stuck with once-a-day backups for I/O intensive workloads—make backups and replicas as often as every 15 minutes!

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-52 Backup from Storage snapshots With the new Veeam 9.5 integration, Veeam can now control the creation of efficient storage-based snapshots on the primary storage array. Veeam controls the initiation of efficient replication to a secondary Nimble array. Only changed compressed blocks are sent across the wire. If encryption is enabled on the data—then it is sent encrypted. Sending data to a more cost-effective Adaptive array allows for more cost effective longer retention cycle on the secondary array.

Secondary Flash Array Use Cases Benefits of HPE Nimble Secondary Flash Array, shown in Figure 10-53, include: Conducting DR testing before a disaster Running Dev/Test based on Production fresh data Test updates before deploying them Troubleshoot issues Train IT staff

||||||||||||||||||||

||||||||||||||||||||

Mitigate Risk

Figure 10-53 Secondary Flash Array Use Cases

Storage Optimizer Storage Optimizer, shown in Figure 10-54, is the single storage optimization solution that combines multiple capabilities focused on cost containment: the visibility needed to reduce redundant or unneeded files; the intelligence needed to store and archive the right files on the right storage tiers; a faster and more efficient way to backup files; and a simpler, smarter way to manage storage infrastructure. It also enables the enterprise to implement a holistic information management and governance strategy—across repositories, across data storage tiers, and across the enterprise.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-54 Storage Optimizer It offers you: Increased cost containment with added intelligence Better infrastructure management Enhanced ability to meet internal objectives Take a giant step forward toward a holistic information governance strategy What is new in Storage Optimizer 5.40: Performance enhancements—Database improvements that dramatically increase indexing throughput, including partitioning, indexing, backup and restore, and security groups. Better extensibility—Support for Netapp (indexing), SharePoint 2016, and Windows Server 2016. Improved Security and Reporting—Metastore security group implementation and HTTPS support, command line support for reporting to Excel tool, Data usage reporting update. Key features include: Analytics-driven: improves data quality and intelligence. Connectivity to a wide range or repositories including File Shares, Hadoop, SharePoint,

||||||||||||||||||||

||||||||||||||||||||

HPE StoreAll, and many others. State-of-the-art scalability that is proven at scales of hundreds of terabytes. Allows you to stub selected documents; move and delete selected data and apply complex policies. Built-in role-based security, review processes, and full auditing. Identifies deduplicated data across different repositories and different storage tiers.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Learning check With Data Protector 10, which component is responsible for ensuring both control and data are secured via the TLS 1.2 channel and guarantees data integrity?

||||||||||||||||||||

||||||||||||||||||||

Compliant archiving challenge Companies are required to understand and comply with many different government and legal regulations that cover records retention, information security, and information privacy. Businesses need to safeguard critical data such as customer information, business correspondence, financial records, and so on. It is necessary to preserve data integrity and be able to access the data quickly when audited or otherwise necessary. Compliance requirements are affecting all industries; Figure 10-55 shows examples of industries that have regulatory needs for data archiving.

Figure 10-55 Which industries have regulatory needs for data archiving? What “compliance” is and what is not? Figure 10-56 shows law vs. policy vs. ethics comparison.

Figure 10-56 Law vs. policy vs. ethics

Technet24

||||||||||||||||||||

||||||||||||||||||||

iTernity Compliant Archive Software Most businesses today are storing and archiving information digitally. Whether tax data, email correspondence, patient or research data, images, or video files: Unstructured, file-based application data use up growing storage space, and have to comply with many regulatory requirements. Figure 10-57 shows key regulations for specific industries.

Figure 10-58 Key regulations for specific industries A trend which is increasing exponentially and which you can only face with effective, intelligent archiving infrastructures. To manage this balancing act, you require highperformance, consolidated, and flexible archiving solutions to reduce to a minimum the business risk of data loss or inadequate proof of justification. Some compliance regulations require data to be archived over long periods of time, and data migration might be needed when the underlying storage technology is upgraded or replaced. Conformity of the HPE iTernity Compliant Archive Software (iCAS) technologies with statutory guidelines is guaranteed at all times. This is because iCAS is assessed and certified by the financial auditing company KPMG, among others. Another advantage of HPE and iCAS, shown in Figure 10-58, is the seamless integration with all storage infrastructures. For long-term availability, it is important that the archived data can be migrated to new technologies as simply and cheaply as possible during ongoing operations.

||||||||||||||||||||

Figure 10-57 iTernity Compliant Archive Software The Content Storage Container (CSC) technologies of the iCAS software store the data and documents to be archived in data containers with the relevant index data, creation date, and retention date. These can be saved to any data medium and still remain verifiable. iCAS is assessed and certified by the financial auditing company KPMG, among others. Seamless integration with all storage infrastructures. CSC technologies store the data and documents to be archived in data containers. –

||||||||||||||||||||

Can be saved to any data medium and still remain verifiable.

HPE Complete—iCAS HPE Complete, shown in Figure 10-59, is a resource enabling the purchase of best in class third-party branded products with the added reliability of Hewlett Packard Enterprise’s interop assurance for a complete validated solution, all via one HPE purchase order.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-59 HPE Complete—iCAS Benefits

One Stop Shop—Purchase complete solutions from HPE that include best in class thirdparty branded products, all on a single HPE purchase order. Validated solution—Confidence of The HPE interop assurance validation for third-party branded products that complement HPE Storage & Server solutions. Customer Experience—Hewlett Packard Enterprise ensures complete customer satisfaction, by troubleshooting the solution and facilitating third-party expertise. For installation, support, and warranty of their products. Unique Value—Minimize risk, complexity, and cost when deploying multi-vendor products and technologies into the HPE ecosystem. The HPE Complete iTernity Compliant Archive Software (iCAS) is a storage middleware that enables compliant archiving by providing WORM functionality, protecting against silent data corruption, and securing long-term data integrity. iCAS

||||||||||||||||||||

||||||||||||||||||||

continually monitors data, uses a self-healing functionality to guarantee the long-term readability and integrity of data, and repairs damaged objects. The system offers unique protection mechanisms to ensure data is also valid and legible. The archive data can also be compressed by up to 50%. This results in better memory usage, reduces energy consumption, and helps customers save costs. HPE StoreEasy Storage with iCAS is an open, flexible, and expandable long-term archiving solution for securing and protecting business data. The solution addresses regulatory compliance requirements, enabling simplified and cost-efficient archiving. The solution is certified by KPMG and complies with the highest security standards. It provides: Functionality for tamperproof archiving by using future-proof HMAC-SHA-512-bit hashes Secure encryption of data Synchronous replication of archive data onto two storage targets or data centers Automatic integrity check and data repair with the iCAS self-healing function

HPE StoreEasy Storage and iCAS This solution complies with the highest security standards and provides: Functionality for tamperproof archiving by using future-proof HMAC-SHA-512-bit hashes Secure encryption of data Synchronous replication of archive data onto two storage targets or data centers Automatic integrity check and data repair with the iCAS self-healing function

Addressing emerging use cases—Compliance and security Enforcement of General Data Protection Regulation (GDPR) begins in May 2018, and penalties are severe: Up to €20 million or 4% of the preceding year’s worldwide turnover. Do not let GDPR compliance slide to the bottom of your priority list. If you reach a point where ransomware affects the personal data you have collected, you need not to only worry about leakage (as in many data breaches), but on how you recover the data to continue your business operations. You will find that a ransomware

Technet24

||||||||||||||||||||

||||||||||||||||||||

infection (or any malware infection) in a considerable number of your workstations and servers that are centric to processing personal data would likely constitute a breach under the GDPR and could trigger the notification obligation in articles 33 and. HPE Format-Preserving Encryption

Organizations seeking GDPR compliance might have stored and processed sensitive PII data within various databases, applications, and systems for several years if not decades. Protecting this data with encryption using traditional techniques results in data incompatible with existing schemas, data structures, and processing requirements. Encrypting structured formatted fields such as customer names, national ID numbers, passport numbers, phone numbers, GPS locations, and dates of birth would require significant database schema and application changes to accommodate the protected data in its new format. Data decryption is then required for each analysis and use, decreasing security overall and imposing additional costs for key management. HPE FPE is a fundamental innovation enabling the HPE SecureData data-centric platform to provide high-strength encryption of data. Technical properties of data encrypted by HPE FPE include: Retain format and structure Retain logical data structure such as checksums, date validity Retain partial nonsensitive values in encrypted fields (partial fields) Retain relationships to other fields and referential integrity where needed Retain the meaning in data and cross-data relationships across records to preserve analytic meaning These properties enable applications, analytic processes, and databases to use the protected data for the majority of use cases, even across distributed systems, platforms, and tools. Protection is applied at the field or partial-field level, leaving nonsensitive portions of fields available for applications while protecting the sensitive parts. HPE FPE can, if required, preserve referential integrity across data sets, so protected data can be consistently referenced and joined. This is especially critical where common identifiers such as phone numbers or IDs are used as references across disparate data sets. HPE FPE adheres to the AES-FF1 per the NIST SP-800-38G FPE standard1 that HPE helped pioneer. This provides enterprises with confidence in the security proofs and standards underpinning HPE FPE. Figure 10-60 shows two emerging cases where compliance during data manipulation is

||||||||||||||||||||

||||||||||||||||||||

important.

Figure 10-60 Addressing emerging use cases—Compliance and security

HPE Enterprise Secure Key Manager Encryption ensures information security and consistency. Data is encrypted at the source, in transit, and at the destination. Modern encryption algorithms are based on digital certificates also known as security keys. Loosing or exposing private keys means losing the information confidentiality; therefore, key management becomes very important. As the number of services within the company increases, the number of security keys that must be managed multiplies. The keys must be kept secure and backed up properly. HPE offers the Enterprise Secure Key Manager (ESKM) appliance to handle backup and restoration of security keys. HPE ESKM, shown in Figure 10-61, provides a centralized key management hardwarebased solution for unifying and automating an organization’s encryption controls by creating, protecting, serving, and auditing access to encryption keys for secure and reliable administration. It is an important product to consider, as it is dangerous to keep the encrypted data and the keys used to encrypt that data on the same appliance and location. Because if that appliance containing encrypted data is compromised, so also are the keys.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-61 HPE Enterprise Secure Key Manager The challenge of protecting large amounts of unstructured data which is dispersed over various geographical locations is a huge undertaking and this dispersed data is best protected by adopting to centralized key management, where keys used for data encryption are managed centrally. Using an HPE ESKM that has automated policies to manage, protect, and serve encryption keys over the life of the data helps to separate keys from the appliance containing encrypted data. Also, it is now mandated for organizations such as Payment Card Industry (PCI) to protect and manage both data and keys under verifiable security controls to pass compliance audits. The HPE StoreOnce security features, when combined with the key management capabilities of HPE ESKM, provide a secure, easy, efficient, and agile data protection solution for both backup data and data in transit. With HPE StoreOnce supported external key management, keys used for encrypting the backup data on StoreOnce are managed by an external HPE ESKM appliance. This arrangement keeps the keys used for encrypting data on HPE StoreOnce on a separate key manager device, thus giving the highest level of data protection. Veeam backup server on HPE Hyper Converged 380 with VMware® vSphere 6.0 Figure 10-62 shows a logical solution diagram for secure data backup using HPE StoreOnce and HPE ESKM. HPE StoreOnce is enrolled as a client to HPE ESKM and data is backed up to HPE StoreOnce from backup servers. HPE StoreOnce uses keys from HPE ESKM to protect the data being stored on HPE StoreOnce.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-62 HPE StoreOnce supports both local and external key management. Keys generated and managed are the Data Encryption Key (DEK), which is used for actual encryption of data on HPE StoreOnce, and the Key Encryption Key (KEK), which is used to encrypt the data encryption key itself. With local key management, the HPE StoreOnce key management system creates, manages, and stores both the data encryption key and the key encryption key. The HPE StoreOnce local keystore manager keeps track of all of the keys used for data encryption. With external key management, HPE StoreOnce is enrolled as a client to an external HPE ESKM. HPE ESKM creates and manages the KEK that is used to encrypt the DEKs on HPE StoreOnce. There is one KEK created per node of HPE StoreOnce on HPE ESKM, and this KEK is also known as the “Master Key” which encrypts and decrypts all the DEKs belonging to the HPE StoreOnce. So, both the StoreOnce OS in local key management mode and HPE ESKM in external key management mode manage and store the KEK, which is used to wrap (encrypt) the DEKs doing the actual encryption of data on HPE StoreOnce as in Figure 10-63.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Figure 10-63 Encyrpting the data with help of ESKM cluster A minimum of two ESKM servers are required for the primary site with the option of scaling up to eight nodes in a cluster. Scaling is generally a business decision, which is generally done to add disaster recovery units, to add robustness to the ESKM cluster, or to have additional fail-over units. An ESKM cluster can support up to 25k clients and 2 million keys.

StoreOnce Security Pack The HPE StoreOnce Security Package, shown in Figure 10-64, delivers a Data at Rest and Data in Flight encryption solution and secure Data Shredding features for data privacy, confidentiality, and integrity of your critical business data while supporting compliance requirements. These are configurable on a by application or by store basis ensuring that you have maximum control over the data you are protecting.

||||||||||||||||||||

||||||||||||||||||||

Figure 10-64 StoreOnce Security Pack Data-at-rest encryption (software encryption)

Supports Catalyst, VTL and NAS AES-256 encryption algorithm and Standard FIPS 140-2 level 1 capable Local Key Management (StoreOnce VSA 2700, 2900, 3100, 3520, 3540, 4500, 4700, 4900, 5100, 5500, 6500, and 6600) Centralized Encrypted Key Management (StoreOnce VSA 3100, 3520, 3540, 5100, 5500, and 6600) Secure Erase functionality (Data Shredding)

Erase data backed up to a VTL, NAS, or Catalyst store NIST SP 800-88 standard Multiple random overwrites (1,3,5, or 7 passes) Data-in-Flight Encryption

Using IPsec Supports:Win2k8 & Win2012 servers, Rhel 5&6 (Red Hat Enterprise Linux) and Sles 11 One way HPE is helping companies with the security of their data protection is with the

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE StoreOnce Security Pack. The HPE StoreOnce Security Pack is a complete and manageable tool that provides the necessary security features that are needed to keep the data protected and secure. HPE StoreOnce Security Pack provides all the HPE StoreOnce security solutions in one license. This makes it easy to install and secure your data backups. HPE StoreOnce Security Packs includes Data-at-Rest Encryption, Secure Erase (also known as Data Shredding), and Data-in-Flight Encryption to ensure you have all the necessary software to keep your data secured and protected. It also provides the necessary protection against unauthorized access to data transferred between StoreOnce Systems including between StoreOnce VSAs. StoreOnce Security Pack licensing options: For single node, one license per System For multinode, one license per couplet and requires all couplets to be licensed

Local key and centralized key management Local use Case

Protection that enables localized key management on the StoreOnce to provide access to business-critical, sensitive, data-at-rest encryption keys. Centralized use Case

Protection that ensures continuous access to business-critical, sensitive data-at-rest encryption keys, both locally and remotely. Benefit—Unified Centralized Encryption Key Management (CEKM) for StoreOnce, StoreEver, 3PAR StoreServ. Rekeying not supported on StoreOnce Systems. Setup and configuration using CLI commands only. You will need to choose local key management or centralized key management. They cannot be used at the same time. Local Solution

||||||||||||||||||||

||||||||||||||||||||

One key per store. Keys are stored in an encrypted form within the StoreOnce System. Will be able to back up keys using the “save the restore configuration option”. Existing non-encrypted stores CANNOT be converted into encrypted stores—Creation of a new store and data copy is required. Random key generator. Keys are stored and generated locally on the StoreOnce System, and cannot be managed outside the StoreOnce System. Centralized Solution

One key per store CEKM solution for generating, storing, serving, controlling, and auditing access to data encryption keys in a secure system StoreOnce (3100, 3520, 3540, 5100, 5500, and 6600) support for ESKM and SafeNet Key Management Interoperability Protocol (KMIP)—compliant encryption key managers Atalla ESKM 4.0 SafeNet KeySecure Manager: www.safenet-inc.com or SafeNet KeySecure Brochure Strong auditable security for encryption keys Secure identity-based access, administration and logging Leveraging the standard “Key Management Interoperability Protocol” (KMIP)

What is ESKM? Enterprise Secure Key Manager (ESKM) is a complete key management solution for generating, storing, serving, controlling, and auditing access to data encryption keys in a secure System. It enables you to protect and ensure continuous access to businesscritical, sensitive, data-at-rest encryption keys, both locally and remotely. ESKM works with HPE ProLiant Servers, 3PAR, NonStop, StoreEver, ESL G3, and more servers and storage devices to protect data at rest. ESKM meets the highest standards for all HPE Storage and any OASIS KMIP enabled solutions.

What is KMIP? Technet24

||||||||||||||||||||

||||||||||||||||||||

The Key Management Interoperability Protocol (KMIP) is an extensible communication protocol that defines message formats for the manipulation of cryptographic keys on a key management server. (March 2018, Wikipedia.org) Long-term strategy of KMIP standards is to allow interoperability of encryption key managers.

||||||||||||||||||||

Learning check What is the minimum number of ESKM servers required for the primary site? A. 1 B. 2 C. 3 D. 4

||||||||||||||||||||

Write a summary of the key concepts presented in this chapter.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Summary HPE BURA solutions provide backup modernization for customers struggling with antiquated backup plans and infrastructure, including comprehensive planning and implementation support. HPE software for BURA includes Data Protector and Big Data Software from Autonomy for an end-to-end data protection solution. The idea behind StoreOnce is to provide users with a single deduplication engine to drive the movement of data across the entire organization. HPE StoreEver Tape portfolio includes tape media, stand-alone tape drives, and enterprise-class tape libraries. HPE offers iCAS technologies to help businesses safeguard critical data such as customer information, business correspondence, and financial records.

||||||||||||||||||||

||||||||||||||||||||

Practice Test Introduction This certification is an important requirement for IT professionals in Presales or Solution Architect roles primarily supporting Enterprise customers. This certification is an important requirement for Solution Architects who need to demonstrate knowledge of Hewlett Packard Enterprise (HPE) Storage solutions. This certification validates that you can identify, recommend, and explain Enterprise Storage Solutions architectures and technologies and translate business requirements into storage solution designs that support applications and data across physical, virtual, and cloud environments with a common architecture and converged management. This certification also validates you can design HPE Backup Solutions including the right Backup, Recovery, and Archive (BURA) strategies for various customer scenarios. This certification demonstrates you can plan, design, validate, and propose hardwareagnostic, Enterprise Storage and Backup Solutions and backup and recovery solutions.

Ideal candidate for this exam The ideal candidate demonstrates the skills typically acquired with one to two years of experience in storage technologies, has completed the HPE-recommended curriculum or has equivalent experience, can assess business requirements, and then develop a storage solution that manages, protects, and optimizes data for enterprise workloads.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Exam details The following are details about the exam: Exam ID: HPE0-J57 Exam Type: Proctored Number of items: 60 Item types: Input text, input numbers, matching, multiple choice (single-response), multiple choice (multiple-response), and point and click Exam time: 1 hour 30 minutes Passing score: 68% Reference material: No online or hard copy reference material will be allowed at the testing site.

HPE0-J57 testing objectives 15%—Foundational storage architectures and technologies 17%—Functions, features, and capabilities of HPE Storage products, solutions, and warranty service offerings 11%—Storage market and competitive opportunities to apply HPE Storage solutions to meet customer needs 27%—Planning, designing, upgrading, and replacing HPE Storage solutions 15%—Performance-tuning, optimizing, and upgrading HPE Storage solutions 15%—Managing, monitoring, administering, and operating HPE Storage solutions

||||||||||||||||||||

Test preparation questions and answers The following questions will help you measure your understanding of the material presented in this book. Read all the choices carefully, as there might be more than one correct answer. Choose all correct answers for each question.

Questions What are the typical disadvantages of traditional storage systems? (Select two). a. Reduced risk exposure b. Complexity c. Rigidity d. Three tiers e. Missing replication Match the HPE converted storage principle to its description: Storage principle Multitenancy Autonomic management

Federated storage

Description Implement self-management and optimization without the need for administrator intervention. Move data between storage systems in a peer-to-peer relationship simply, dynamically, and nondisruptively to improve storage utilization, balance workloads, and ease storage migration. Meet service-level agreements (SLAs) and deliver virtual private arrays for secure workload segregation.

Which operating system is preconfigured on the HPE StoreEasy 3000 Gateway? a. Red Hat Enterprise Linux b. SUSE Linux Enterprise Server c. Windows Storage Server d. Windows Server Datacenter edition

||||||||||||||||||||

Which functionality enables the HPE 3PAR StoreServ system to perform subvolume Technet24

||||||||||||||||||||

data movement? a. Adaptive Optimization b. Dynamic Optimization c. Autonomic Rebalance d. System Tuner A customer wants to implement a Fibre Channel (FC)-based storage area network (SAN) on their campus. They are concerned about the security of the FC traffic between switches in different buildings. a. enable in-flight encryption on the ISLs (inter-switch links) b. set the default zoning to ”noaccess” c. create a special zone that consists only of ISL ports d. use ICLs (inter-chassis links) instead of ISLs to connect the switches You use the HPE Renew Program to deliver an HPE 3PAR StoreServ demonstration unit to a potential customer who needs a block storage system to run an OLTP (online transaction processing) application. The customer estimates at least 20TB of usable capacity and wants to use the existing 10Gb/s integrated Small Computer System Interface (iSCSI) infrastructure. Low latency is important, but cost is also a concern. What should you include in the demonstration unit design to deliver the proof of concept? a. (8) 2TB NL (nearline) drives, a two-port 10Gb/s Ethernet adapter, enable File Persona b. (8) 1.92TB solid-state drives (SSDs), (16) 1.2TB SAS (serial attached SCSI) drives, two-port 10Gb/s iSCSI/FCoE (Fibre Channel over Ethernet) adapter c. (16) 920GB MLC (multi-level cell) SSDs, two-port 10Gb/s iSCSI/FCoE adapter d. (8) 400GB SSDs, (16) 1.2TB SAS drives, a two-port 8Gb/s FC HBA (host bus adapter). Why should customers use HPE StoreVirtual DSM (device-specific module) for Windows? a. It provides improved performance architecture over native MPIO solutions. b. It automatically adjusts transfer block size. c. It automatically adjusts the available bandwidth between the switch and the array.

||||||||||||||||||||

||||||||||||||||||||

d. It automatically combines HPE thin provisioning and volume migration technology. Which storage technology allows a customer to provision a group of servers with 20TB of storage from only 10TB of actual physical storage? a. deduplication b. space reclamation c. compression d. thin provisioning A customer’s environment consists of four fabrics with four HPE B-series, fixed-port, SAN switches in each fabric. Which tool is recommended to manage and monitor this environment? a. HPE SAN Design Reference Guide b. HPE OneView c. HPE InfoSight d. HPE SAN Network Advisor Which HPE management tool provides a single, integrated management environment for a converged infrastructure and provides storage provisioning functions? a. HPE 3PAR SSMC b. HPE Storage Operations Manager c. HPE OneView d. HPE InfoSight for HPE 3PAR You are demonstrating an HPE Synergy, HPE C-series FC switches, and HPE StoreVirtual integration. Which objects need to be configured in HPE OneView to provision host access to volumes on the storage system? (Select two.) a. Volume b. Volume template c. Server profile d. Server profile template e. Firmware bundle

||||||||||||||||||||

While demonstrating an HPE 3PAR StoreServ 20000 R2 array, the customer asks you to Technet24

||||||||||||||||||||

demonstrate the File Persona features. Which objects do you need to create to demonstrate these features? (Select two.) a. Virtual file server b. Storage container c. Storage Volume d. Remote copy group e. File share Match the incident severity level to its definition: Incident severity level Definition Severity 1 Severity 2 Severity 3 Severity 4

Production environment severely impaired, interrupted, or compromised. Production environment or system is down or at severe risk. No business or user impact. Nonproduction system is down or degraded.

Customer is concerned about the availability of the storage system in case of a hardware failure. Which support level will provide a defined onsite response time for hardware failures, that cannot be resolved remotely, during weekends and holidays at the lowest costs? a. Foundation Care Next Business Day b. Foundation Care 24x7 c. Proactive Care d. Proactive Care Advanced 24x7 What would you do to implement the recommended cache size for HPE SmartArray SmartCache? a. Set legacy cache to 50% read/write to enable write-back support. b. Enable each logical disk volume to use its own cache volume. c. Upgrade the controller with 4GB of FBWC (Flash Backed Write Cache) memory.

||||||||||||||||||||

||||||||||||||||||||

d. Provide 1GB of metadata space for every TB of accelerator space. What are advantages of using SSDs over performance HDDs? (Select three.) a. higher random IOPS b. lower write latency c. more secure with self-encryption d. increased power consumption e. higher ambient temperature support f. less power consumption Which Redundant Array of Inexpensive Disk (RAID) is configured by default on HPE Nimble Storage arrays? a. Triple parity with possibility to modify to RAID 5. b. Triple parity with possibility to modify to RAID 6. c. Triple parity and cannot be modified. d. RAID 6 with possibility to modify to triple parity. A customer is interested in a converged networking and storage connectivity for HPE BladeSystem solution and directly attached HPE 3PAR StoreServ array. Customer requires redundant connectivity to interconnect modules and at least 4Gb connectivity for storage and 10Gb for network traffic. Which Virtual Connect module can a solution architect recommend? a. VC FlexFabric 20/40 F8 b. VC FlexFabric 10/24 c. VC Flex-10/10D d. FC 16Gb Which typical building blocks are used to design HPE 3PAR StoreServ 9450 solution? (Select three). a. Base enclosure with two controller nodes b. Base enclosure with four controller nodes and 24 internal drives c. Base enclosure with eight controller nodes d. SFF (small form factor) Drive enclosure e. LFF (large form factor) Drive enclosure

||||||||||||||||||||

Technet24

||||||||||||||||||||

f. Additional HBAs An SMB (small and medium business segment) customer is considering a consolidation of few applications to an entry-level array with good price/performance ratio. Which HPE Storage product can be the first pick to be introduced by solution architect? a. HPE XP7 b. HPE MSA c. HPE 3PAR StoreServ d. HPE Synergy Which stage of the typical consulting engagement process usually contains a research of the company and decision of proper person to be contacted? a. Prepare b. Interview c. Plan d. Propose

||||||||||||||||||||

Match the HPE resource to its description: Resource or tool

Description

HPE Storage Information Library SPOCK (Single Point of Connectivity Knowledge) HPE Demonstration Portal HPE QuickSpecs

Showcases how HPE technologies lead, innovate, and transform businesses by providing a central location for all demonstrations, webinars, and supporting collateral. Convenient central resource providing technical overviews and specifications for HPE hardware and software. Provides access to installation guides, configuration guides, user guides, references such as release notes, planning manuals, and service and maintenance guides. Portal used to obtain detailed information about supported HPE Storage product and hardware configurations.

A customer wants to disable short-time peaks from the alerting in the threshold alert functionality of the HPE 3PAR StoreServ Management Console. Which threshold alert settings can be adjusted to achieve this?

||||||||||||||||||||

a. Change the sampling to low res b. Object in HPE 3PAR IMC (Inform Management Console) c. Change the logical metrics operator to OR (Logical disjunction) d. Change the sampling to hourly A customer is considering an integration between Intranet portal and HPE 3PAR StoreServ. The goal is to display a hardware and capacity status of the array on the Intranet homepage. Which interface is the most appropriate for such tasks? a. SOAP interface to HPE 3PAR StoreServ Management Console b. RESTful API c. ESKM (Enterprise Secure Key Manager) d. Service Processor (physical) Which operating system can be used for Nimble Setup Manager to discover Nimble array? a. Nimble Setup Manger is OS agnostic. b. Windows c. Red Hat Enterprise Linux d. VMware ESX 6.5 U1 and newer Which interfaces are commonly used in HPE StoreOnce environments? (Select two.) a. NTFS (New Technology File System) b. EXT4 c. EXT3 d. NFS (Network File System) e. CIFS (Common Internet File System) What are the supported capacity licenses for HPE StoreOnce VSA (virtual storage appliance)? (Select three.) a. 1TB b. 2TB c. 32TB d. 36TB

||||||||||||||||||||

Technet24

||||||||||||||||||||

e. 50TB f. 100TB A customer finished the implementation of an HPE StoreVirtual system. When validating the installed storage arrays and connecting several volumes to a VMware farm, the customer reveals that a current backup solution for protecting virtual machines does not provide satisfactory RTO (Recovery Time Objective). Which additional HPE solutions can be demonstrated to fill the gap in the customer’s current environment (Select two)? a. HPE Recovery Manager Central for VMware b. HPE OneView c. HPE 3PAR StoreServ Management Console d. HPE 3PAR Inform Management Console e. HPE StoreOnce A customer is considering a partial backup during week days. Environment is highly dynamic, with high percentage of daily changes and small backup window. Longer downtime during restore is acceptable. Which backup can a solution architect recommend during the week? a. Differential b. Incremental c. Copy d. Full

||||||||||||||||||||

Match the backup-related factor to its description: Term

Description

RTO (Recovery Time Objective) Retention

How much data is allowed to go unprotected and how far back in time data must be recovered?.

RPO (Recovery Point Objective)

How long the customer is willing to wait for the data to be recovered and the maximum allowable downtime. How long the data needs to be kept available. Can range from seconds to decades, depending on company policies and government regulations.

||||||||||||||||||||

What is the benefit of using HPE StoreOnce as a target for Veeam Backup & Replication solution? a. Veeam can use NFS for source side deduplication. b. Veeam can use CIFS for source side deduplication. c. Veeam can use Catalyst for source side deduplication. d. Veeam can use Catalyst and StoreOnce VSA without a capacity limitation.

||||||||||||||||||||

Technet24

||||||||||||||||||||

Practice Test Answers and explanations What are the typical disadvantages of traditional storage systems? (Select two.) a. Reduced risk exposure b. Complexity c. Rigidity d. Three tiers e. Missing replication

||||||||||||||||||||

Answers B and C are correct. IT environments with a large number of different legacy storage systems and architectures are complex, costly, and rigid. These types of environments cannot meet customer needs for primary storage. Answers A, D, and E are incorrect. Reduced risk exposure is not a disadvantage, three tiers are used frequently in latest arrays, and many traditional storage systems are also capable of replication. For more information, see Chapter 1: HPE Storage enterprise market. Match the Hewlett Packard Enterprise (HPE) converted storage principle to its description: Storage principle Autonomic management Federated storage

Multitenancy

Description Implement self-management and optimization without the need for administrator intervention. Move data between storage systems in a peer-to-peer relationship simply, dynamically, and nondisruptively to improve storage utilization, balance workloads, and ease storage migration. Meet service-level agreements (SLAs) and deliver virtual private arrays for secure workload segregation.

||||||||||||||||||||

Correct matching is shown in the table above. For more information, see Chapter 1: HPE Storage enterprise market. Which operating system is preconfigured on the HPE StoreEasy 3000 Gateway? a. Red Hat Enterprise Linux b. SUSE Linux Enterprise Server c. Windows Storage Server d. Windows Server Datacenter edition C is correct. The HPE StoreEasy 3000 Gateway comes with preconfigured hardware and either Microsoft Windows Storage Server 2016 or 2012 R2. A, B, and D are incorrect. StoreEasy 1000 and 3000 products are not using Linux or Windows Server Datacenter edition. For more information, see Chapter 2: HPE Storage portfolio. Which functionality enables the HPE 3PAR StoreServ system to perform subvolume data movement? a. Adaptive Optimization b. Dynamic Optimization c. Autonomic Rebalance d. System Tuner

||||||||||||||||||||

A is correct. Adaptive Optimization uses the subvolume data movement engine built into HPE 3PAR OS to relocate subvolume regions to the fastest storage tier available. It also moves less active regions to slower tiers to help ensure space availability for newly active regions. B, C, and D are incorrect. HPE 3PAR Dynamic Optimization is moving full LUNs (logical unit numbers, representing volumes) between tiers. HP 3PAR Autonomic Rebalance optimizes the use of future capacity expansions without requiring administrator intervention. It analyzes how volumes on the HP 3PAR Storage System are using physical disk space and makes intelligent, autonomic adjustments to ensure optimal volume distribution when new hardware is added to the system. System Tuner maintains peak HP 3PAR StoreServ Storage performance by autonomically and nondisruptively detecting and resolving performance bottlenecks and storage hotspots. For more information, see Chapter 2: HPE Storage portfolio. A customer wants to implement a Fibre Channel (FC)-based storage area network

Technet24

||||||||||||||||||||

(SAN) on their campus. They are concerned about the security of the FC traffic between switches in different buildings. Which recommendation can storage architect provide to the customer? a. enable in-flight encryption on the ISLs (inter-switch links) b. set the default zoning to “noaccess” c. create a special zone that consists only of ISL ports d. use ICLs instead of ISLs to connect the switches A is correct. In-flight encryption on the ISLs provides security for frames while they are passing between two switches. B, C, and D are incorrect. Zoning is not protecting the data once it leaves the switch. ICLs are increasing bandwidth, but they do not provide anything from the security prospective. For more information, see Chapter 2: HPE Storage portfolio and SAN Design Reference Guide. You use the HPE Renew Program to deliver an HPE 3PAR StoreServ demonstration unit to a potential customer who needs a block storage system to run an OLTP application. The customer estimates at least 20TB of usable capacity and wants to use the existing 10Gb/s integrated Small Computer System Interface (iSCSI) infrastructure. Low latency is important, but cost is also a concern. What should you include in the demonstration unit design to deliver the proof of concept? a. (8) 2TB NL drives, a two-port 10Gb/s Ethernet adapter, enable File Persona b. (8) 1.92TB SSDs (solid-state drives), (16) 1.2TB SAS (serial attached SCSI) drives, two-port 10Gb/s iSCSI/FCoE (Fibre Channel over Ethernet) adapter c. (16) 920GB MLC SSDs, two-port 10Gb/s iSCSI/FCoE adapter d. (8) 400GB SSDs, (16) 1.2TB SAS drives, a two-port 8Gb/s FC HBAs (host bus adapters)

||||||||||||||||||||

B is correct. The combination provides both required capacity and connectivity. A, C, and D are incorrect. B and C do not provide required capacity; D does not provide required connectivity. For more information, see Chapter 2: HPE Storage portfolio. Why should customers use HPE StoreVirtual DSM (device-specific module) for

||||||||||||||||||||

Windows? a. It provides improved performance architecture over native MPIO (Multipath I/O) solutions b. It automatically adjusts transfer block size. c. It automatically adjusts the available bandwidth between the switch and the array. d. It automatically combines HPE thin provisioning and volume migration technology. A is correct. HPE StoreVirtual DSM for Microsoft MPIO provides enhanced MPIO functionality, such as automatic creation of an IO path to each storage system in the cluster on which the volume resides and improved performance architecture over native MPIO solutions: B, C, and D are incorrect. HPE StoreVirtual DSM does not adjust transfer block size or available bandwidth. HPE StoreVirtual DSM has nothing to do with thin provisioning or volume migration. For more information, see Chapter 2: HPE Storage portfolio. Which storage technology allows a customer to provision a group of servers with 20TB of storage from only 10TB of actual physical storage? a. deduplication b. space reclamation c. compression d. thin provisioning D is correct. Thin provisioning allows creating volumes larger than actually available resources. A, B, and C are incorrect. Space reclamation, compression, or deduplication will not allow provisioning of more storage than available. For more information, see Chapter 3: HPE virtualization, fabrics, and converged management. A customer’s environment consists of four fabrics with four HPE B-series, fixed-port, SAN switches in each fabric. Which tool is recommended to manage and monitor this environment? a. HPE SAN Design Reference Guide b. HPE OneView

||||||||||||||||||||

Technet24

||||||||||||||||||||

c. HPE InfoSight d. HPE SAN Network Advisor D is correct. HPE B-series SAN Network Advisor Software provides comprehensive management of data center fabrics, including configuration, monitoring, and management of all B-series Directors, Switches, and HBAs. A, B, and C are incorrect. SAN Design Reference Guide is not a management tool. HPE OneView is not a monitoring tool for switches, even it is capable of doing automated zoning. HPE InfoSight is not providing monitoring and managing capabilities for Bseries FC SAN switches. For more information, see Chapter 3: HPE virtualization, fabrics, and converged management. Which HPE management tool provides a single, integrated management environment for a converged infrastructure and provides storage provisioning functions? a. HPE 3PAR StoreServ Management Console (SSMC) b. HPE Storage Operations Manager c. HPE OneView d. HPE InfoSight for HPE 3PAR C is correct. HPE OneView is a converged infrastructure management tool for managing servers, storage, and networking resources. A, B, and D are incorrect. HPE 3PAR SSMC can be used only for managing HPE 3PAR StoreServ and HPE Storage Operations Manager can be also used only for storage management. HPE InfoSight for HPE 3PAR is a predictive analytics system predicting and preventing problems; it does not provide storage provisioning functions. For more information, see Chapter 3: HPE virtualization, fabrics, and converged management. You are demonstrating an HPE Synergy, HPE C-series FC switches, and HPE StoreVirtual integration. Which objects need to be configured in HPE OneView to provision host access to volumes on the storage system? (Select two.) a. Volume b. Volume template c. Server profile

||||||||||||||||||||

||||||||||||||||||||

d. Server profile template e. Firmware bundle Answers A and C are correct; server profile is a configuration object that contains most of the server configuration, and volume is part of this configuration. Answers B, D, and E are incorrect. Volume template does not represent a volume that can be attached to the server; it does not exist at the array level. Server profile template cannot be assigned to a physical server. Firmware bundle has no active role in volume vs. server attachment. For more information, see Chapter 3: HPE virtualization, fabrics, and converged management. While demonstrating an HPE 3PAR StoreServ 20000 R2 array, the customer asks you to demonstrate the File Persona features. Which objects do you need to create to demonstrate these features? (Select two.) a. Virtual file server b. Storage container c. Storage Volume d. Remote copy group e. File share

||||||||||||||||||||

A and E are correct. File Persona features is using File stores, FPGs (File Provisioning Groups), File shares, Virtual file servers, and CPGs objects. B, C, and D are incorrect. Storage container is not a 3PAR configuration object. Volumes and remote copy groups are used by block persona. For more information, see Chapter 2: HPE Storage portfolio and Chapter 3: HPE virtualization, fabrics, and converged management. Match the incident severity level to its definition:

Technet24

||||||||||||||||||||

Incident severity level Definition Severity 1 Severity 2 Severity 3 Severity 4

Production environment or system is down or at severe risk. Production environment severely impaired, interrupted, or compromised. Nonproduction system is down or degraded. No business or user impact.

Correct matching is shown in the table above. Severity 1 incident level is ”critical down”, level 2 means “Critically degraded,” severity 3 is “Normal,” and severity level 4 means “Low”. For more information, see Chapter 4: HPE Storage services. Customer is concerned about the availability of the storage system in case of a hardware failure. Which support level will provide a defined onsite response time for hardware failures that cannot be resolved remotely, during weekends and holidays at the lowest costs? a. Foundation Care Next Business Day b. Foundation Care 24x7 c. Proactive Care d. Proactive Care Advanced 24x7 Answer B is correct. Foundation Care 24x7 includes four-hour onsite response for incidents with covered hardware that cannot be resolved remotely; service is available 24 hours per day, seven days per week including holidays observed by HPE. Answers A, C, and D are incorrect. For Foundation Care Next Business Day hardware and software support, service is available nine hours per day, between 8:00 a.m. and 5:00 p.m. in the customer’s time zone, Monday through Friday, excluding holidays observed by HPE. Proactive Care is available at higher price point comparing to Foundation Care. For more information, see Chapter 4: HPE Storage services. What would you do to implement the recommended cache size for HPE SmartArray SmartCache? a. Set legacy cache to 50% read/write to enable write-back support.

||||||||||||||||||||

||||||||||||||||||||

b. Enable each logical disk volume to use its own cache volume. c. Upgrade the controller with 4GB of FBWC (Flash-Based Write Cache) memory. d. Provide 1GB of metadata space for every TB of accelerator space. Answer D is correct. The SmartCache solution consumes a portion of the FBWC memory module on the Smart Array controller for metadata. To ensure sufficient storage for accelerator and metadata, you should recommend 1 or 2GB of FBWC memory or 1GB of metadata space for every TB of accelerator space. Answers A, B, and C are incorrect. 4GB of FWBC memory is not needed; 1 or 2GB is sufficient. Logical disk volumes are using dedicated cache volumes; however, this is not connected to size settings. When using SmartCache, you should set the legacy cache for 100% write operation, not 50% read/write ratio. For more information, see Chapter 5: Planning HPE Storage solutions. What are advantages of using SSDs over performance HDDs? (Select three.) a. higher random IOPS b. lower write latency c. more secure with self-encryption d. increased power consumption e. higher ambient temperature support f. less power consumption

||||||||||||||||||||

Answers A, B, and F are correct. HPE SSDs provide significantly better random read and write IOPS compared to HDDs. Although sequential read and write throughput is also improved over HDDs, the greatest benefit is recognized in random data applications. SSDs use flash memory technology. They have no moving parts and do not experience the latency or synchronization issues that are drawbacks of HDDs. SSDsustained data transfer rates are much faster than those of HDDs. They also consume much less power. Answers C, D, and E are incorrect. Self-encrypted HDDs are relatively common, and SSDs have much lower consumption than HDDs. HPE does not set higher ambient temperatures support for SSDs comparing to HDDs; on the contrary, NVMe (NonVolatile Memory Expresss) SSDs are not supported for some ASHRAE A3 and A4 compliant configurations. For more information, see Chapter 5: Planning HPE Storage solutions. Which Redundant Array of Inexpensive Disk (RAID) is configured by default on HPE Technet24

||||||||||||||||||||

Nimble Storage arrays? a. Triple parity with possibility to modify to RAID 5. b. Triple parity with possibility to modify to RAID 6. c. Triple parity and cannot be modified. d. RAID 6 with possibility to modify to triple parity. Answer C is correct; answers A, B, and D are incorrect. The RAID is preconfigured as triple parity and cannot be modified on HPE Nimble Storage arrays. For more information, see Chapter 5: Planning HPE Storage solutions. A customer is interested in a converged networking and storage connectivity for HPE BladeSystem solution and directly attached HPE 3PAR StoreServ array. The Customer requires redundant connectivity to interconnect modules and at least 4Gb connectivity for storage and 10Gb for network traffic. Which Virtual Connect module can a solution architect recommend? a. VC FlexFabric 20/40 F8 b. VC FlexFabric 10/24 c. VC Flex-10/10D d. FC Fibre Channel 16Gb Answer A is correct; answers B, C, and D are incorrect. Converged storage and networking connectivity together with direct connection to HPE 3PAR StoreServ is supported on Virtual Connect FlexFabric 10/24 and 20/40 F8 modules; however, required bandwidth cannot be achieved using VC FlexFabric 10/24. For more information, see Chapter 5: Planning HPE Storage solutions. Which typical building blocks are used to design HPE 3PAR StoreServ 9450 solution? (Select three.) a. Base enclosure with 2 controller nodes b. Base enclosure with 4 controller nodes and 24 internal drives c. Base enclosure with 8 controller nodes d. SFF (small form factor) Drive enclosure e. LFF (large form factor) Drive enclosure f. Additional HBAs

||||||||||||||||||||

||||||||||||||||||||

Answers A, D, and F are correct. The backplane of the 9450 is 8U and can contain either two or four controller nodes; the base model can be selected, and then two nodes can be added as necessary. Next host facing HBAs should be added to each node pair. Since the 9450 can have only SSDs, only SFF enclosures can be added. Answers B, C, and E are incorrect. The controllers on the 9450 do not contain any drives; drives can only be added to drive enclosures. LFF enclosures are not supported, as 9450 requires SSDs. 9450 does not support eight controller nodes, but two or four controller nodes. For more information, see Chapter 5: Planning HPE Storage solutions. An SMB (Small and Medium Business segment) customer is considering a consolidation of few applications to an entry-level array with good price/performance ratio. Which HPE storage product can be the first pick to be introduced by solution architect? a. HPE XP7 b. HPE MSA c. HPE 3PAR StoreServ d. HPE Synergy Answer B is correct; answers A, C, and D are incorrect. It is important to align the right solution by workload and use case. For small deployments, consolidating a few applications with good price/performance, but limited SLA requirements, MSA2x00 is a great start. For databases, emerging applications, Storage-as-a-Service with VMs (virtual machines), Microsoft applications, or container environments, start with Nimble and consider 3PAR for greater configurability, performance, and multitenant, multi-array scale. For a large private cloud, performance databases, HPE Synergyattach, ITaaS, VMs, and OLTP analytics, start with 3PAR and consider XP7 for extreme availability and/or mainframe connectivity. HPE Synergy is rather a converged infrastructure and server product than storage. For more information, see Chapter 6: Staging an effective storage consulting engagement. Which stage of the typical consulting engagement process usually contains a research of the company and decision of proper person to be contacted? a. Prepare b. Interview c. Plan

||||||||||||||||||||

Technet24

||||||||||||||||||||

d. Propose Answer A is correct. There are five stages to the typical consulting process. Prepare, interview, plan, propose, and present. “Prepare” stage typical includes learning as much as possible before beginning to engage a customer and researching the company and the people to find out about areas of concern, and deciding, who is the best person to contact. Answers B, C, and D are not correct. Interview, plan, and propose phase are used in later time during typical consulting process, when researching of the company needs to be finished already as well as selecting appropriate people for further contact. For more information, see Chapter 6: Staging an effective storage consulting engagement. Match the HPE resource to its description: Resource or tool

Description

HPE Demonstration Portal HPE QuickSpecs

Showcases how HPE technologies lead, innovate, and transform businesses by providing a central location for all demonstrations, webinars, and supporting collateral. Convenient central resource providing technical overviews and specifications for HPE hardware and software. Provides access to installation guides, configuration guides, user guides, references such as release notes, planning manuals, and service and maintenance guides. Portal used to obtain detailed information about supported HPE Storage product and hardware configurations.

HPE Storage Information Library SPOCK (Single Point of Connectivity Knowledge)

Correct matching is shown in the table above. For more information, see Chapter 6: Staging an effective storage consulting engagement. A customer wants to disable short-time peaks from the alerting in the threshold alert functionality of the HPE 3PAR StoreServ Management Console. Which threshold alert settings can be adjusted to achieve this? a. Change the sampling to low res

||||||||||||||||||||

||||||||||||||||||||

b. Object in HPE 3PAR IMC (Inform Management Console) c. Change the logical metrics operator to OR d. Change the sampling to hourly Answer D is correct; answers A, B, and C are incorrect. Results can be viewed at high resolution (five-minute sample), hourly for medium resolution, and daily for low resolution. For more information, see Chapter 7: Administering and monitoring HPE Storage solutions. A customer is considering an integration between Intranet portal and HPE 3PAR StoreServ. The goal is to display a hardware and capacity status of the array on the Intranet homepage. Which interface is the most appropriate for such tasks? a. SOAP (Simple Object Access Protocol) interface to HPE 3PAR StoreServ Management Console b. RESTful API c. ESKM (Enterprise Secure Key Manager) d. Service Processor (physical) Answer B is correct. RESTful API is ideal for integration of various products. Answers A, C, and D are incorrect. SSMC does not have SOAP interface. ESKM and service processors are not used for hardware and capacity monitoring. For more information, see Chapter 7: Administering and monitoring HPE Storage solutions. Which operating system can be used for Nimble Setup Manager to discover Nimble array? a. Nimble Setup Manger is OS agnostic b. Windows c. Red Hat Enterprise Linux d. VMware ESX 6.5 U1 and newer

||||||||||||||||||||

Answer B is correct. Initial setup for administration is completed through the Nimble Setup Manager, which is a Windows Utility that uses automatic array discovery. Answers A, C, and D are incorrect. Nimble Setup Manager only supports Windows; it is neither OS agnostic nor it supports Linux or VMware. For more information, see Chapter 7: Administering and monitoring HPE Storage Technet24

||||||||||||||||||||

solutions. Which interfaces are commonly used in HPE StoreOnce environments? (Select two.) a. NTFS (New Technology File System) b. EXT4 c. EXT3 d. NFS (Network File System) e. CIFS (Common Internet File System) Answers D and E are correct; answers A, B, and C are incorrect. HPE StoreOnce is not using NTFS, EXT4, or EXT3 interfaces. HPE StoreOnce allows a selection of the backup device type that best fits customer needs and is supported by backup software provider. Available interfaces include VTL (virtual tape library), Network-Attached Storage (NAS—CIFS and NFS), and StoreOnce Catalyst. For more information, see Chapter 8: Upgrading, optimizing, and tuning HPE Storage solutions. What are the supported capacity licenses for HPE StoreOnce VSA? (Select three.) a. 1TB b. 2TB c. 32TB d. 36TB e. 50 TB f. 100TB

||||||||||||||||||||

Answers A, C, and E are correct. StoreOnce VSA provides the flexibility to deploy as per data protection requirements, with licensing options available at 4TB, 10TB, 20TB, 32TB, and 50TB capacity points. To enable scalability, a 4TB license can be upgraded to 10TB, 10TB to 20TB, 20TB to 32TB, and 32TB to 50TB. Answers B, D, and F are incorrect; there are no licenses for 2TB, 36TB, and 100TB capacities. For more information, see Chapter 9: HPE backup solutions. A customer finished the implementation of an HPE StoreVirtual system. When validating the installed storage arrays and connecting several volumes to a VMware farm, the customer reveals that a current backup solution for protecting virtual machines does not

||||||||||||||||||||

provide satisfactory RTO (Recovery Time Objective). Which additional HPE solutions can be demonstrated to fill the gap in the customer’s current environment (Select two)? a. HPE Recovery Manager Central for VMware b. HPE OneView c. HPE 3PAR StoreServ Management Console d. HPE 3PAR Inform Management Console e. HPE StoreOnce Answers A and E are correct. RMC-V together with HPE StoreOnce can provide a great backup solution to StoreVirtual and VMware environment. Answers B, C, and D are incorrect. HPE OneView is not providing backup capabilities for StoreVirtual storage arrays. HPE 3PAR Management Consoles (SSMC and IMC) cannot manage or back up StoreVirtual environment. For more information, see Chapter 9: HPE backup solutions. A customer is considering a partial backup during week days. Environment is highly dynamic, with high percentage of daily changes and small backup window. Longer downtime during restore is acceptable. Which backup can a solution architect recommend during the week? a. Differential b. Incremental c. Copy d. Full

||||||||||||||||||||

Answer B is correct. The advantages of the incremental backup are that the size of backed up data and the backup window is small each day. The disadvantage is that each day must be restored separately, consuming time. Answers A, C, and D are incorrect. Differential backup offers faster restore and longer backup window, which is not suitable for scenario. Full and copy backups are not partial backups. For more information, see Chapter 10: Designing HPE BURA solutions. Match the backup-related factor to its description:

Technet24

||||||||||||||||||||

Term

Description

RPO

How much data is allowed to go unprotected and how far back in time data must be recovered? RPO can be measured in days, hours, or even minutes. How long the customer is willing to wait for the data to be recovered and the maximum allowable downtime. How long the data needs to be kept available. Retention time can range from seconds to decades, depending on company policies and government regulations.

RTO Retention

Correct matching is shown in the table above. For more information, see Chapter 10: Designing HPE BURA solutions. What is the benefit of using HPE StoreOnce as a target for Veeam Backup & Replication solution? a. Veeam can use NFS for source-side deduplication. b. Veeam can use CIFS for source-side deduplication. c. Veeam can use Catalyst for source-side deduplication. d. Veeam can use Catalyst and StoreOnce VSA without a capacity limitation.

||||||||||||||||||||

Answer C is correct. HPE StoreOnce can be used as a Veeam Backup & Replication target in Backup over LAN or LAN-free backup. Veeam can use Catalyst for sourceside deduplication over FC. Answer A, B, and C are incorrect. HPE StoreOnce NAS targets (NFS and CIFS) are not offering source-side deduplication. StoreOnce VSA capacity is not unlimited. For more information, see Chapter 10: Designing HPE BURA solutions.

||||||||||||||||||||

||||||||||||||||||||

Learning Check Answers

Technet24

||||||||||||||||||||

||||||||||||||||||||

Chapter 1 The following questions will help you to measure your understanding of the material presented in this chapter. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. To survive in today’s fast-paced enterprise environment, businesses need to become more agile. What are the three key components fundamental to Hewlett Packard Enterprise (HPE) Composable Infrastructure that enable agility? Fluid resource pools Software-defined intelligence Unified API (Application Programming Interface) What is provided by HPE OneView for a Composable Infrastructure? Software-Defined Intelligence Name two key data center trend that customers are addressing today. All Flash systems Software-defined storage Name two key data center trends that customers are addressing today. Get more value for better decisions Lower costs and boost agility Save time and reduce complexity Write a summary of the key concepts you learned in this module. HPE Storage meets the needs of the Idea Economy. HPE solutions provide stepping stones from a traditional to a Composable Infrastructure. To keep up with storage trends, business is moving toward automated costoptimized solutions. HPE Composable Infrastructure is designed around fluid resource pools, softwaredefined intelligence, and a unified API. HPE Synergy allows businesses to reduce CapEx with improved efficient economics.

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

Chapter 2 The following questions will help you to measure your understanding of the material presented in this chapter. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. An SMB (small and medium business segment) customer has an entry-level storage area network (SAN), and they ask you about archiving. What products would you discuss with them and why? MSL tape libraries offer unattended backup and archiving capability for entrylevel SAN customers. They provide excellent scalability, density, and performance. Your customer wants to add more capacity to his StoreVirtual 3200. The StoreVirtual 3200 supports both SFF (small form factor) and LFF (large form factor) drive enclosures. What can you tell them about the maximum number of disk slots available in each enclosure? SFF: 25 LFF: 12 Flash is all about consolidating racks and racks of storage. You need density to do that, and we are already pushing the boundaries there by being first to market with larger and larger solid-state drives (SSDs). Right now, you can do over 15 TB on a single SSD, but if you want to continue to grow this density, eventually you would sacrifice performance for more capacity. There are two new flash technologies that will help us achieve this. Can you name them? Storage class memory Nonvolatile Memory express Which requirements indicate that you should propose an HPE 3PAR StoreServ 9450 rather than an 8450? (Select three.) a. Future growth requirements b. Limited budget c. High resiliency requirements d. High IOPS (Input/output operations per second) requirement e. Requirement for multiple data services such as File Persona and replication

||||||||||||||||||||

||||||||||||||||||||

f. Gen5 ASIC capabilities required g. General purpose file requirements What two Express features are available with HPE Recovery Manager Central Software (RMC)? Express Protect Express Restore What are the primary features of HPE 3PAR OS software? (Select three.) a. Workloads can move between members of a federation to rebalance storage resources dynamically. b. Autonomic provisioning features eliminate traditional storage planning. c. Advanced virtualization capabilities deliver high and predictable service levels. d. Data protection and copy space reclamation features improve capacity utilization. e. Streaming offers optimal disk and bandwidth utilization and provides cost-effective replication over any distance.

||||||||||||||||||||

Name two management tools used to configure HPE 3PAR StoreServ. SSMC (StoreServ Management Console) CLI HPE InfoSight is a Reactive Analytic tool. False Name the two product families in the HPE primary storage portfolio. StoreServ Nimble Storage StoreVirtual Your customer has a 3PAR StoreServ 8000 and wants to transition to the new allinclusive license. What minimum OS version should they be operating on? Any 3.2.2 (All single-system software except Online Import and RMC App Suite) 3.2.2 MU3 (All single-system software) 3.3.1 (All single-system software)

Technet24

||||||||||||||||||||

||||||||||||||||||||

Write a summary of the key concepts presented in this module. HPE provides a comprehensive portfolio of block, file, and object-level storage. Management tools enhance visibility and assist in maintaining updates. HPE provides simplified all-inclusive licensing along with O/S and bundled software titles. Nondisruptive Federation improves efficiencies for migration and life cycle management.

||||||||||||||||||||

||||||||||||||||||||

Chapter 3 The following questions will help you to measure your understanding of the material presented in this module. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. Describe software-defined storage. Software-defined storage is fully virtualized with no physical hardware dependency for provisioning, performance, optimization, tenants, and so on. What information is visualized in the TreeMap tool? What needs to be registered to provide vCenter credentials? Datastore TreeMap is a virtualization tool to show the IO and latency of datastores and VMs. You must have vCenter Plugin registered. What Protocol does not support a flat SAN? FCoE (Fibre Channel over Ethernet) Name the three approaches to an HPE SAN Design. An HPE standard SAN fabric topology design A subset or variation of an HPE standard SAN fabric topology design A custom SAN fabric topology design What are the three main components of the HPE OneView domain? Logical interconnect groups Uplink sets Networks What components are included within the hyperconverged systems? Servers, storage switch, highly available shared storage, SSD arrays, backup, WAN optimization, and data protection apps Write a summary of the key concepts presented in this module. Virtualization is the pooling of multiple network storage devices into what appears to be a single storage device that is managed from a central console. HPE Virtual Connect for BladeSystem provides simplified storage, server, and network management in virtualized environments. Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE SAN Fabric topologies support a range of high-availability solutions, including interoperability in heterogeneous networks. With HPE OneView, you can view and manage your storage system and storage pools, add existing volumes, create new volumes, and create volume templates. HPE Hyper Converged system is an entire data center in a box. HPE Composable Infrastructure is designed around fluid resource pools, softwaredefined intelligence, and a unified API.

||||||||||||||||||||

||||||||||||||||||||

Chapter 4 The following questions will help you to measure your understanding of the material presented in this module. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. In the HPE Partner Ready Program, there are six competencies. List them below. Cloud Automation Software-Defined Infrastructure Data Analytics Infrastructure Object Storage Business Continuity & Data Protection Workplace Experience Which resources can Foundation Care, Proactive Care, and Proactive Care Advanced customers access through the HPE Support Center? Published proactive care reports Subscription-based, hardware-related proactive service notifications Support forums Expanded web-based searches of entitled technical support documents Certain HPE proprietary service diagnostic tools A web-based tool for submitting questions directly to HPE HPE and available third-party hosted knowledge databases Service tool, which is used to browse, select, and schedule services (using credits) Write a summary of the key concepts presented in this module. The HPE Partner Ready Delivery Services Program supports partners who sell or deliver HPE Services. The program includes designing and deploying new storage solutions, managing customer storage environments, and providing responsive support. Comprehensive support services from HPE include Datacenter Care, Proactive Care Services, Foundation Care, and Financial Services. HPE Solution Consulting Services span several areas, including analytics, data

Technet24

||||||||||||||||||||

||||||||||||||||||||

management, applications, data centers, workloads, and cloud. HPE Financial Services help technology users in every customer segment make the most of their IT investment. HPE Pointnext provides services and consulting for the digital transformation age.

||||||||||||||||||||

||||||||||||||||||||

Chapter 5 The following questions will help you to measure your understanding of the material presented in this module. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. When discussing concerns around life expectancy of SSDs, what reassurance can you offer regarding HPE SSDs? All HPE SSDs now come with a seven-year warranty. Can frequently accessed data read from any disk tier to dynamic random access memory (DRAM) be moved into AFC? Not data read from an SSD tier. StoreFabric M-Series switch performance scales from 1Gbe to 100Gbe. True When discussing response times with your customer, which solution would typically offer the lowest: integrated Small Computer System Interface (iSCSI) or Fibre Channel? From a response time standpoint, iSCSI has some additional overhead vs. FC, and so you can expect iSCSI response time to be about 1.6 times the equivalent FC response time. What is the minimum operating system version required for deduplication on Nimble Hybrid flash systems? Version 5.0 Write a summary of the key concepts presented in this module. Planning and designing an enterprise storage system that can handle large volumes of data and a large number of users is a multistep process with many decision points along the way. The Redundant Array of Inexpensive Disk (RAID) configuration you choose for the storage system depends on the customer’s plans for data fault tolerance, data availability, and capacity growth. SANs provide flexibility in system management, configuration, connectivity, and performance to meet the needs of the changing business environment. When performing an initial configuration of a StoreServ 8000 storage system, there are certain factors to consider such as RAID configurations that permit highly available cage.

Technet24

||||||||||||||||||||

||||||||||||||||||||

The key difference between Flex-10 and FlexFabric is that FlexFabric modules leverage the built-in converged network adapter in BladeSystem servers to provide FCoE or iSCSI connectivity. HPE Reference Architectures include IT building blocks that are assembled to address key enterprise workloads.

||||||||||||||||||||

||||||||||||||||||||

Chapter 6 The following questions will help to measure your understanding of the materials presented in this module. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. You are looking to book a meeting to discuss a project on which you are working. With whom should you be discussing business goals and objectives? CEO From where can you download the storage sizing tools? Storage Assessment Foundry Æ Sizing Tools To obtain hands-on access in the Nimble Live Demo portal, how many days’ lead time do you need to wait? Two days Name the two types of tests that are required for a POC (proof of concept). Resilience tests Functionality tests Write a summary of the key concepts presented in this module. HPE presales consultants and partners need to learn as much as possible about the customer’s business situation and needs to convey value to each customer. HPE presales consultants and partners need to learn as much as possible about the customer’s business situation and needs to convey value to each customer. HPE presales consultants and partners need to learn as much as possible about the customer’s business situation and needs to convey value to each customer. HPE offers a full spectrum of online resources to help you work with customers and implement successful storage solutions.

Technet24

||||||||||||||||||||

||||||||||||||||||||

Chapter 7 The following questions will help you to measure your understanding of the material presented in this module. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. There are a number of ways to manage a 3PAR StoreServ array. Name four options. SSMC CLI OneView API SMI-S (Storage Management Initiative Specification) The Excel add-in for SSMC supports Excel 2013 and 2016. True What are the two tabs in StoreOnce Manager graphical user interface (GUI) for updating firmware? Server Storage The initial Nimble Array setup must be completed in the CLI. False HPE OneView supports KVM (KVM— Kernel-based Virtual Machine) True Write a summary of the key concepts presented in this module. HPE 3PAR manageability and monitoring is provided through many different options. Managing and monitoring StoreOnce from the management software can be completed through the GUI or the CLI. Nimble Setup Manager is a Windows Utility that uses automatic array discovery. HPE offers a wide range of monitoring tools to help customers improve IT processes, such as Analytics for HPE OneView, HPE OneView for VMware vCenter, and HPE OneView for Microsoft System Center.

||||||||||||||||||||

||||||||||||||||||||

Technet24

||||||||||||||||||||

||||||||||||||||||||

Chapter 8 The following questions will help you to measure your understanding of the material presented in this module. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. You are looking to add some new drives in your 3PAR StoreServ 8400 array. To maintain best practices, what would be the minimum upgrade drive quantity? 4 When you are about to perform an online controller node software upgrade, what hostbased software needs to be active? Multipathing IO What is a good practice to perform before upgrading Basic Input/Output System (BIOS)? Enable Remote Event Suppression mode The Adaptive Flash Cache simulator can only be run in the CLI. False Tunesys phase 3 detects if new disks were added and will automatically redistribute the data across them. False Upgrading the service processor pre-operating system 3.3.1 uses the Service Console. False Write a summary of the key concepts presented in this module. When upgrading storage components, it is necessary to adhere to best practices. Configurations should be optimized for cost, availability, and performance. Storage should be tuned to maintain balanced system utilization. 3PAR StoreServ operating system can be upgraded via SSMC and CLI. Service Processor can be updated using the new service console.

||||||||||||||||||||

Chapter 9 The following questions will help you to measure your understanding of the material presented in this module. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. Object storage is best suited for: a. Storing and sharing files that might be accessed on a frequent basis b. Databases and other business application c. Large amounts of infrequently accessed and long-term unstructured data Low-bandwidth deduplication is: a. Server side b. Target side Peer Copy is supported with RMC 4.1 and greater. Which features are supported? (Select two.) a. Homogenous replication between StoreVirtual VSA and 3PAR b. Heterogeneous replication between StoreVirtual VSA and 3PAR c. Unidirectional replication d. Bidirectional replication

||||||||||||||||||||

The StoreOnce 3100 is supported with Cloud Bank. False Existing MSL Tape Drives and Tape Media can be used in the MSL6480. True The HPE StoreOnce Get Protected Guarantee program is for StoreOnce systems, but not StoreOnce VSA. False Write a summary of the key concepts presented in this module. HPE StoreOnce products support entry-level to high-end backup and recovery through a simple standardized interface. HPE StoreOnce provides a simple, consistent, high-performing architecture that spans the organization.

Technet24

||||||||||||||||||||

||||||||||||||||||||

HPE StoreOnce RMC serves as the single integration and control point for data protection. HPE StoreOnce Cloud Bank Storage leverages object storage for long-term retention of backup data copies. HPE StoreEver Tape portfolio includes tape media, standalone tape drives, and enterprise-class tape libraries.

||||||||||||||||||||

Chapter 10 The following questions will help you to measure your understanding of the material presented in this module. Read all the choices carefully because there might be more than one correct answer. Select or write the correct answers for each question. Match each severity level with its definition.

Which backup model turns off the archive bit? (Select two.) a. Full backup b. Differential backup c. Incremental backup

||||||||||||||||||||

What three steps help in defining a backup strategy? Define the requirements and constraints for backups. Understand the factors that influence the backup solution. Technet24

||||||||||||||||||||

Prepare the backup design. When sizing the RMC host, how many vCPUs do you need to support Peer Copy and Express Protect? a. 1 b. 2 c. 3 d. 4 With Data Protector 10, which component is responsible for ensuring both control and data are secured via the TLS 1.2 channel and guarantees data integrity? Cell Manager What is the minimum number of Enterprise Secure Key Manager (ESKM) servers required for the primary site? a. 1 b. 2 c. 3 d. 4

||||||||||||||||||||

Write a summary of the key concepts presented in this module. HPE BURA solutions provide backup modernization for customers struggling with antiquated backup plans and infrastructure, including comprehensive planning and implementation support. HPE software for BURA includes Data Protector and Big Data Software from Autonomy for an end-to-end data protection solution. The idea behind StoreOnce is to provide users with a single deduplication engine to drive the movement of data across the entire organization. HPE StoreEver Tape portfolio includes tape media, standalone tape drives, and enterprise-class tape libraries. HPE offers iCAS (iTernity Compliant Archive Software) technologies to help businesses safeguard critical data such as customer information, business correspondence, and financial records.

||||||||||||||||||||

||||||||||||||||||||

Technet24