Transcription
Building a Modern Data CenterPrinciples and Strategies of DesignWritten by Scott D. Lowe, James Green and David DavisIn partnership with
BUILDING A MODERN DATA CENTERPrinciples and Strategies of DesignWritten byScott D. Lowe, James Green and David DavisPartners, ActualTech Media
Building a Modern Data CenterPrinciples and Strategies ofDesignAuthor:Editor:Scott D. Lowe, ActualTech MediaJames Green, ActualTech MediaDavid Davis, ActualTech MediaHilary Kirchner, Dream Write CreativeCover Design:Atlantis ComputingLayout:Braeden Black, Avalon Media ProductionsProject Manager: Geordie Carswell, ActualTech MediaCopyright 2016 by Atlantis ComputingAll rights reserved. This book or any portion there of may not be reproduced or usedin any manner whatsoever without the express written permission of the publisherexcept for the use of brief quotations in a book review.Printed in the United States of AmericaFirst Printing, 2016ISBN 978-1-943952-07-6ActualTech MediaOkatie Village Ste 103-157Bluffton, SC 29909USAwww.actualtechmedia.com
Table of ContentsForeword . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiiiChapter 1:Introduction . . . . . . . . . . . . . . . . . . . . . . . . . .14IT Is Changing . . . and It Must . . . . . . . . . . . . . . . . . . . 14Simplification Is the New Black . . . . . . . . . . . . . . . . . .15Focus on Cost Reduction . . . . . . . . . . . . . . . . . . . . .17Focus on Customer Service and the Business . . . . . . . . . .19Adapt or Die . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20Flattening of the IT Organization . . . . . . . . . . . . . . . . .21IT as an Operational Expense . . . . . . . . . . . . . . . . . . . 22Chapter 2:Data Center Evolution . . . . . . . . . . . . . . . . . . . . . 25A History of the Modern Data Center . . . . . . . . . . . . . . . 26The Rise of the Monolithic Storage Array . . . . . . . . . . . . . 27The Virtualization of Compute — Software Defined Servers . . . . 33The No-Spin Zone: The Move from Disk to Flash . . . . . . . . . 38The Fall of the Monolithic Storage Array . . . . . . . . . . . . .41The Emergence of Convergence . . . . . . . . . . . . . . . . .44The Role of Cloud . . . . . . . . . . . . . . . . . . . . . . . . . 47Cloud Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47Cloud Drivers . . . . . . . . . . . . . . . . . . . . . . . . . . . 49Chapter 3:Emerging Data Center Trends . . . . . . . . . . . . . . . . . 53The Emergence of SDDC . . . . . . . . . . . . . . . . . . . . . 53Commoditization of Hardware . . . . . . . . . . . . . . . . . . . 54Shift to Software Defined Compute . . . . . . . . . . . . . . . . 56Shift to Software Defined Storage . . . . . . . . . . . . . . . . . 57Shift to Software Defined Networking . . . . . . . . . . . . . . . 59Shift to Software Defined Security . . . . . . . . . . . . . . . . 60The Parallel Paths of SDS and Hyperconvergence . . . . . . . . 61iv
The Details of SDS . . . . . . . . . . . . . . . . . . . . . . . . . 62What Is Hyperconverged Infrastructure? . . . . . . . . . . . . . 65The Relationship Between SDS and HCI . . . . . . . . . . . . . 68The Role of Flash in Hyperconvergence . . . . . . . . . . . . . . 69Where Are We Now? . . . . . . . . . . . . . . . . . . . . . . . . 71Chapter 4:Modern IT Business Requirements . . . . . . . . . . . . . . 76The Business Requirement Challenge . . . . . . . . . . . . . . . 76Risk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78Assurance, Business Continuity, and Disaster Recovery . . . . . 80The Regulatory Landscape . . . . . . . . . . . . . . . . . . . . . 82Avoiding Lock-In (Hypervisor, Storage, Server) . . . . . . . . . . 83Changing the Perception of IT . . . . . . . . . . . . . . . . . . . 84Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86Changing Budgetary Landscape . . . . . . . . . . . . . . . . . . 86The Changing Career Landscape . . . . . . . . . . . . . . . . . 88Complexity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90Agility and Performance . . . . . . . . . . . . . . . . . . . . . . 92Automation and Orchestration . . . . . . . . . . . . . . . . . . . 93Self-Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95The Data Growth Challenge . . . . . . . . . . . . . . . . . . . . 97Resiliency and Availability . . . . . . . . . . . . . . . . . . . . . 98Chapter 5:Making Your Data Center Agile: Principles & Strategies . . 101Think Big . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102Start Small . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104Move Fast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106Chapter 6:Transforming Your Data Center . . . . . . . . . . . . . . . .113Align the Data Center and Business Needs . . . . . . . . . . . .114Where to Address the Low-Hanging Fruit . . . . . . . . . . . . .115Test/Dev . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .117Remote/Branch Offices . . . . . . . . . . . . . . . . . . . . . .118v
Server Virtualization . . . . . . . . . . . . . . . . . . . . . . . 120Big Data/Analytics . . . . . . . . . . . . . . . . . . . . . . . . .121Disaster Recovery . . . . . . . . . . . . . . . . . . . . . . . . 123Virtual Desktop Infrastructure . . . . . . . . . . . . . . . . . . 124How to Adobt SDDC . . . . . . . . . . . . . . . . . . . . . . . 125Software Defined Compute . . . . . . . . . . . . . . . . . . . 126Software Defined Storage . . . . . . . . . . . . . . . . . . . . 128Software Defined Networking . . . . . . . . . . . . . . . . . . 130Simplification . . . . . . . . . . . . . . . . . . . . . . . . . . . 131Eliminate Monolithic Storage Systems . . . . . . . . . . . . . . 132Implement Opportunistic Hyperconvergence . . . . . . . . . . 134Management . . . . . . . . . . . . . . . . . . . . . . . . . . . 135Full Stack Management . . . . . . . . . . . . . . . . . . . . . 135Key Features . . . . . . . . . . . . . . . . . . . . . . . . . . . 136Chapter 7:Software Defined Storage . . . . . . . . . . . . . . . . . . .141The SDS Landscape . . . . . . . . . . . . . . . . . . . . . . . 142What Is Software Defined Storage? . . . . . . . . . . . . . . . 142What Isn’t Software Defined Storage? . . . . . . . . . . . . . . 144SDS Compared to Traditional Storage . . . . . . . . . . . . . . 144SDS Requirements . . . . . . . . . . . . . . . . . . . . . . . . 146Abstraction, Pooling, and Storage Virtualization . . . . . . . . . 146Advanced Data Services . . . . . . . . . . . . . . . . . . . . . .151Resilience/Data Protection . . . . . . . . . . . . . . . . . . . . 160Chapter 8:Hyperconverged Infrastructure . . . . . . . . . . . . . . . 165Hyperconvergence Defined . . . . . . . . . . . . . . . . . . . . 166Hyperconvergence Implementation Options . . . . . . . . . . . 168SDS in Hyperconvergence . . . . . . . . . . . . . . . . . . . . 168The Hyperconvergence Design Model . . . . . . . . . . . . . . 169Virtual Storage Appliances . . . . . . . . . . . . . . . . . . . . 171Appliance vs. Software/Reference Architecture . . . . . . . . . .174Hypervisor Choice . . . . . . . . . . . . . . . . . . . . . . . . 176vi
Server Hardware Choice . . . . . . . . . . . . . . . . . . . . . 177The Question of Scale . . . . . . . . . . . . . . . . . . . . . . 178Scale Up Defined . . . . . . . . . . . . . . . . . . . . . . . . . 178Scale Out (Linear Scalability) Defined . . . . . . . . . . . . . . 179Do You Have Scale Up and Scale Out Options? . . . . . . . . . 179Chapter 9:From the Field: Software DefinedStorage & Hyperconverged Infrastructure 2016 . . . . . . 182Technology Domain Knowledge . . . . . . . . . . . . . . . . . 184Virtualization Penetration . . . . . . . . . . . . . . . . . . . . . 185Hypervisor Usage . . . . . . . . . . . . . . . . . . . . . . . . . 187Storage Capacity . . . . . . . . . . . . . . . . . . . . . . . . . .191Data Growth . . . . . . . . . . . . . . . . . . . . . . . . . . . 195Storage Performance . . . . . . . . . . . . . . . . . . . . . . . 198Current Storage Systems . . . . . . . . . . . . . . . . . . . . . 199The Form Factor Debate . . . . . . . . . . . . . . . . . . . . . 202Business Workload Support . . . . . . . . . . . . . . . . . . . 203Remote Office and Branch Office Support . . . . . . . . . . . . 204Flash Storage Deployment . . . . . . . . . . . . . . . . . . . . 205Software Defined Storage and HyperconvergedInfrastructure Deployment Intent . . . . . . . . . . . . . . . . . 206Software Defined Storage and HyperconvergedInfrastructure Deployment Experience . . . . . . . . . . . . . . 207Hyperconverged Infrastructure orSoftware Defined Storage Deployment Timeframe . . . . . . . . 208Software Defined Storage andHyperconverged Infrastructure Decision Criteria . . . . . . . . . 209Chapter 10:IT Transformation . . . . . . . . . . . . . . . . . . . . . . .211The Challenge . . . . . . . . . . . . . . . . . . . . . . . . . . . 212The Solutions . . . . . . . . . . . . . . . . . . . . . . . . . . . 213Automation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213Orchestration . . . . . . . . . . . . . . . . . . . . . . . . . . . 216DevOps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219vii
Industry Needs . . . . . . . . . . . . . . . . . . . . . . . . . . 222Open Standards and Interoperability . . . . . . . . . . . . . . . 224Transparency on Pricing and Cost . . . . . . . . . . . . . . . . 227Performance Benchmarking Standards . . . . . . . . . . . . . 231Chapter 11:The Future of the Data Center . . . . . . . . . . . . . . . 235Future Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . 236Containers Instead of Virtual Machines . . . . . . . . . . . . . . 237Open Source Tools . . . . . . . . . . . . . . . . . . . . . . . . 242Flash Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . 243Non-Volatile Memory Express . . . . . . . . . . . . . . . . . . 246Beyond Today’s Flash . . . . . . . . . . . . . . . . . . . . . . 247Pooling of Resources . . . . . . . . . . . . . . . . . . . . . . 250The Cloud . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252viii
About the AuthorsScott D. Lowe,Partner, ActualTech MediaScott Lowe is a vExpert and partner and Co-Founder ofActualTech Media. Scott has been in the IT field for closeto twenty years and spent ten of those years in filling theCIO role for various organizations. Scott has writtenthousands of articles and blog postings and regularlycontributes to www.EnterpriseStorageGuide.com &www.ActualTech.io.James Green,Partner, ActualTech MediaJames Green is an independent blogger atwww.virtadmin.com, two-time vExpert, serial Tech FieldDay delegate, and works as a virtualization consultant inthe Midwest.David M. Davis,Global Trainer & vExpertDavid Davis is a partner and co-founder of ActualTechMedia. With over 20 years in enterprise technology, hehas served as an IT Manager and has authored hundredsof papers, ebooks, and video training courses. He’s a 6x vExpert, VCP, VCAP, & CCIE# 9369. You’ll find hisvSphere video training at www.Pluralsight.com and heblogs at www.VirtualizationSoftware.com andwww.ActualTech.io.
About ActualTech MediaActualTech Media provides enterprise IT decision makers with theinformation they need to make informed, strategic decisions as theymodernize and optimize their IT operations.Leading 3rd party IT industry influencers Scott D. Lowe, David M.Davis and special technical partners cover hot topics from the softwaredefined data center to hyperconvergenceand virtuali
The modern data center is an exciting place, and it looks nothing like the data center of only 10 years past. The IT industry and the world in general are changing at an exponential pace. In fact, according to Moore’s Law (named after the co-founder of Intel, Gordon Moore), computing power doubles every few years. Due to the limitations of the laws of physics, this pace has slowed a bit over .
