VIrtualization 2.0 for Dummies

8/10/2019 VIrtualization 2.0 for Dummies

1/88

Hanish RathodJosh Townsend

The next generation ofvirtualization technologies

Operations management

and virtualized storage and

network capabilities

Technical tips, advice

and best practicesBrought to you by:

Learn about:

Virtualization

2.0

VMwareSpecialEditionMakingEverythi

ngEasier!


2/88

VMware (NYSE:VMW), a global leader invirtualization and cloud infrastructure, delivers

customer-proven solutions that accelerate IT by

reducing complexity and enabling more flexible,

agile service delivery. VMware enables enterprises

to adopt a cloud model that addresses their uniquebusiness challenges. VMwares approach accelerates

the transition to cloud computing while preserving

existing investments and improving security and

control. With more than 500,000 customers and55,000 partners, VMware solutions help

organizations of all sizes lower costs, increase

business agility and ensure freedom of choice.

For more information, visit www.vmware.com.

About VMware Inc.


3/88

By Hanish Rathod andJosh Townsend

Virtualization

2.0


4/88

Virtualization 2.0 For Dummies

Published byJohn Wiley & Sons, LtdThe AtriumSouthern GateChichester

West SussexPO19 8SQEngland

For details on how to create a customFor Dummiesbook for your business ororganisation, contact [email protected] . For informationabout licensing theFor Dummiesbrand for products or services, contactBrandedRights&[email protected] .

Visit our Home Page on www.customdummies.com

Copyright 2014 by John Wiley & Sons Ltd, Chichester, West Sussex, England

All Rights Reserved. No part of this publication may be reproduced, stored in aretrieval system or transmitted in any form or by any means, electronic, mechani-cal, photocopying, recording, scanning or otherwise, except under the terms of theCopyright, Designs and Patents Act 1988 or under the terms of a licence issued bythe Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London, W1T 4LP,UK, without the permission in writing of the Publisher. Requests to the Publisherfor permission should be addressed to the Permissions Department, John Wiley &Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, England,or emailed [email protected],or faxed to (44) 1243 770620.

Trademarks:Wiley, the Wiley Publishing logo, For Dummies, the Dummies Manlogo, A Reference for the Rest of Us!, The Dummies Way, Dummies Daily, The Funand Easy Way, Dummies.com and related trade dress are trademarks or registeredtrademarks of John Wiley & Sons, Inc. and/or its affiliates in the United States andother countries, and may not be used without written permission. All other trade-marks are the property of their respective owners. Wiley Publishing, Inc., is notassociated with any product or vendor mentioned in this book.

LIMIT OF LIABILITY/DISCLAIMER OF WARRANTY: THE PUBLISHER, THE AUTHOR, ANDANYONE ELSE INVOLVED IN PREPARING THIS WORK MAKE NO REPRESENTATIONS ORWARRANTIES WITH RESPECT TO THE ACCURACY OR COMPLETENESS OF THE CONTENTS OFTHIS WORK AND SPECIFICALLY DISCLAIM ALL WARRANTIES, INCLUDING WITHOUTLIMITATION WARRANTIES OF FITNESS FOR A PARTICULAR PURPOSE. NO WARRANTY MAY BECREATED OR EXTENDED BY SALES OR PROMOTIONAL MATERIALS. THE ADVICE ANDSTRATEGIES CONTAINED HEREIN MAY NOT BE SUITABLE FOR EVERY SITUATION. THIS WORKIS SOLD WITH THE UNDERSTANDING THAT THE PUBLISHER IS NOT ENGAGED IN RENDERINGLEGAL, ACCOUNTING, OR OTHER PROFESSIONAL SERVICES. IF PROFESSIONAL ASSISTANCEIS REQUIRED, THE SERVICES OF A COMPETENT PROFESSIONAL PERSON SHOULD BE SOUGHT.NEITHER THE PUBLISHER NOR THE AUTHOR SHALL BE LIABLE FOR DAMAGES ARISINGHEREFROM. THE FACT THAT AN ORGANIZATION OR WEBSITE IS REFERRED TO IN THIS WORKAS A CITATION AND/OR A POTENTIAL SOURCE OF FURTHER INFORMATION DOES NOT MEANTHAT THE AUTHOR OR THE PUBLISHER ENDORSES THE INFORMATION THE ORGANIZATIONOR WEBSITE MAY PROVIDE OR RECOMMENDATIONS IT MAY MAKE. FURTHER, READERS

SHOULD BE AWARE THAT INTERNET WEBSITES LISTED IN THIS WORK MAY HAVE CHANGEDOR DISAPPEARED BETWEEN WHEN THIS WORK WAS WRITTEN AND WHEN IT IS READ.

Wiley also publishes its books in a variety of electronic formats. Some content thatappears in print may not be available in electronic books.

ISBN: 978-1-119-02432-3
mailto:[email protected]:BrandedRights&[email protected]:BrandedRights&[email protected]://www.customdummies.com/mailto:[email protected]:[email protected]:[email protected]:[email protected]://www.customdummies.com/mailto:BrandedRights&[email protected]:[email protected]


5/88

Contents

Introduction ...............................................................................1

About This Book ......................................................................2

Foolish Assumptions ..............................................................2

How This Book Is Organized .................................................2

Icons Used in This Book ........................................................3Where to Go from Here...........................................................4

Chapter 1: The What and Why of Virtualization .............5

In the Beginning: The Mighty Mainframe ...........................6

Trend Spotting: Data, Data Everywhere ............................7

Seven Good Reasons to Virtualize .......................................8

Chapter 2: The Evolution of Virtualization ......................11

Hypervisor 101: Sharing the Love ......................................11

Todays Challenges ................................................................12

The Next Generation: Virtualization 2.0............................13

Chapter 3:

Virtualization 2.0:Starting with Operations Management ..........................17

Capacity Management ..........................................................18

Capacity management: What is it? ..............................18

How its done: Millennium Pharmacyoptimized IT resources with VMware ......................19

Performance Monitoring ......................................................22Performance monitoring: What is it? .........................22

How its done: Cornerstone optimizedIT performance with VMware ....................................23


6/88

Chapter 4: Virtualization 2.0: Deep Dive ..........................27

Operations Managementfor Virtualized Environments ...........................................28

The challenge: Improve efficiency, and fast ...............29

How it works ....................................................................29

Virtualized Backup ................................................................31

The goal: Simplify, simplify ...........................................31

How it works ....................................................................32

Virtualized Storage ................................................................34

The answer: Virtual storage-area networks ..............34

How it works ....................................................................35

Virtualized Networking .........................................................36

The mission: Make life easier .......................................37

How it works ....................................................................38

Chapter 5: Real-world Outcomesof Virtualization 2.0 ................................................................41

IT Outcome: Data Center Virtualizationand Standardization ...........................................................42

IT Outcome: Streamlined and AutomatedData Center Operations ....................................................43

IT Outcome: Security ControlsNative to Infrastructure .....................................................45

IT Outcome: High Availabilityand Resilient Infrastructure ..............................................47

IT Outcome: App and InfrastructureDelivery Automation ..........................................................50

Chapter 6: Virtualization 2.0 Preflight ..............................53Preparing to Move from Physical to Virtual ....................53

Using Traditional vs. Virtual Storage .................................55


7/88

Sizing and Managing Shared Storage.................................56

Addressing Security and Compliance ...............................58

Using Operations Management

to Meet Business Objectives ............................................59Adding Operations Management .......................................59

Prepare Carefully to Optimize Resource Utilization ......60

Consider the Appropriate Permissionsfor Different User Types ....................................................60

Model Your Virtual Environment

on Your Business Structure ..............................................61Create Policies for Efficient Resource Management ......61

Configure Badges for Quick Health Checks .....................62

Identify the Needs of Workgroupsto Configure Capacity Settings ........................................62

Choose How You Want to be Alerted ................................62

Set up Email Notifications for Administrative Alerts .....63Schedule Reports to Help Address Bottlenecks

Before They Occur .............................................................63

Unify Your View of the Virtual Environment ....................64

Chapter 7: Ten Best Practices for Virtualization 2.0 ...65

Prepare your Physical Servers ...........................................65

Install Your Hypervisor ........................................................66

Configure Back-end Storage Devices Correctly ..............67

Evaluate and Redesign the Networkbefore Virtualizing ..............................................................68

Secure Your Environment ....................................................69

Performance-tune for Latency-sensitive Workloads ......70

Enable Advanced Featuresfor Performance and Availability ....................................71

Optimize VMs So Theyre Stable and Efficient ................72


8/88

Monitor and Optimize Continuously .................................73

Spread the Word ....................................................................74

Chapter 8: Ten Useful Resourcesfor Virtualization 2.0 ..............................................................75

About the Authors ...........................................................78

Acknowledgements .........................................................78


9/88

Introduction

Welcome to Virtualization 2.0For Dummies, yourguide to understanding and managing that big,

borderless frontier known as the virtualized datacenter.

Its a world that knows no boundaries. A world wheredata, devices and user expectations are growing likemad. Where businesses are expected to operate 24x7while running faster, leaner, more securely and moreprofitably each year.

If youre in IT, lets face it, youre an enabler. Yourecreating the virtualized environment that makes theseoutsized expectations possible. And, ultimately, youreresponsible for controlling what youve created. In thewords of Doctor Frankenstein, It is aliiive. And thats agood thing, as long as it can be controlled andmanaged.

This book can help. It provides tips, insights and

advice for gaining the upper hand on your virtualizeddata center. With a quick read, you can betterunderstand how to manage the virtual environmentyoure responsible for, and begin using thetechnologies of Virtualization 2.0 to bring the fullbenefits and cost savings of virtualization to yourorganization.


10/88

2

About This BookThis book is packed with information for understandingand managing a virtualized data center, including howvirtualization got here, why virtualization matters, andwhat you need to know before deploying or upgradinga virtualized environment.

Foolish AssumptionsIn writing this book, weve made some assumptionsabout you. We assume that:

You work in IT or within an IT organization.

You understand the basics of virtualization andhave already virtualized some aspects of your

data center. You believe that virtualization is a good thing but

you need ideas for how to make it better.

How This Book Is OrganizedVirtualization 2.0 For Dummiesis divided into eight

chapters: Chapter 1: The What and Why of Virtualization.

What it is, why it matters, and why virtualizeddata centers are all the rage in this ever-changingdigital world.

Chapter 2: The Evolution of Virtualization.How we got here, the challenges we face, andwhat the future holds.

Chapter 3: Virtualization 2.0 Starting withOperations Management. The key to seeing and


11/88

3

managing a sprawling mix of virtual machines andassets, and keeping your data center humming.

Chapter 4: Virtualization 2.0 Deep Dive. Meet thefour horsemen of Virtualization 2.0 backup,storage, the network, and at the foundation,operations management and see how theywork to expand virtualization beyond thecompute layer.

Chapter 5: Real-world Outcomes.Learn how fiveIT outcomes can be achieved in the real worldwith Virtualization 2.0.

Chapter 6: Preflight. Things to consider,roadblocks to avoid, and what to expect beforeyou launch.

Chapter 7: Best Practices.Wondering how others

have gone about this? Check out these technicaltips, techniques and insights.

Chapter 8: Useful Resources. Ready to make themove to Virtualization 2.0? Here are our picks forthe top resources as you expand virtualizationbeyond your servers.

Icons Used in This BookThe Dummies man spotlights real-life exam-ples to illustrate a point.

The knotted string highlights importantinformation to bear in mind.


12/88

4

Follow the target for tips and top-notchadvice from the trenches.

Watch out for these potential pitfalls.

Where to Go from HereYou can take the traditional route and read this bookstraight through. Or you can skip between sections,using the section headings as your guide to pinpointthe information you need. Whichever way you choose,

you cant go wrong. Both paths lead to the sameoutcome a better understanding of Virtualization 2.0and how it can make your data center even moreefficient.


13/88

Chapter 1

The What and Why ofVirtualization

In This Chapter Virtualization defined

How it all began

Drivers and trends

Why data centers go virtual

Theres this scene inMatrix Reloadedwhere Neo isattacked by an endlessly propagating pack of Mr. Smith

clones. Ten Smiths become hundreds, then thousands.They just keep coming, relentless and identical.

Its a cool CGI effect. And not a bad metaphor for virtualiz-ation. In fact, if you take away the sunglasses and slow-mobody slams, youve got the basic idea behind managing avirtual data center: Endless numbers of virtual machines,created in software, cloning fast, and coming at you fromevery direction.

If you manage a virtualized data center, you understand.According toStrategic Benchmarks 2014: ServerVirtualization, a March 2014 Forrester Research, Inc. report,over 50 per cent of todays x86 workloads were virtualized in


14/88

6

2013. By 2015, respondents expect to virtualize more than 70per cent of their x86 servers. Virtualization has clearly gone

mainstream in the 21st Century, but why? How did we reacha point where virtual machines and resources areeverywhere, propagating like mad?

In the Beginning: The Mighty MainframeThe data center as we know it today evolved from the

mainframes of the 1950s. Back then, mainframes werehoused in large, climate-controlled, secure facilities withsophisticated power and cooling systems. Mainframes arecomplex, expensive and powerful, and after 50 years, theystill play a role in todays computing hierarchy.

By the 1990s, many mainframes were being replaced byserver rooms where banks of servers are connected,

powered, cooled and maintained onsite. Server roomsbrought improvements over mainframes, allowing a moremodular approach to provisioning resources andhandling increased data growth. Like mainframes, serverrooms are complex and expensive, and can generateenough heat to warm an entire building.

In the 2000s, offsite colocation facilities entered the scene,

allowing companies to house their servers and data inmultiple locations. It was the beginning of the moderndata center: Data started flowing to servers that may havebeen offsite or onsite, creating a level of abstractionbetween a company and its data. That trend continuestoday with cloud-connected, virtualized data centers.

Like the early mainframes, todays virtualized data centercomes in all shapes and sizes. In fact, its misleading torefer to the data center as though a single blueprintexists. It doesnt.


15/88

7

Some are fully in-sourced, built from the ground up withonsite servers, applications, network infrastructure and

IT teams to manage everything in-house. Others are fullyoutsourced, using a combination of cloud platforms toconnect a business to its data located in distant facilities.Others fall somewhere in between, a hybrid combinationof cloud-based services and on-premises infrastructureand servers. Regardless of the approach, the commonground is that a virtualized data center is designed to

evolve with a business. It supports a level of growth andIT flexibility that used to be inconceivable.

Trend Spotting: Data, Data EverywhereBusinesses around the world are using that new-foundflexibility to grow their data stores as never before. In its

Global Cloud Index, Cisco predicts that the worlds datacenters will be handling 7.7 zettabytes of data per year by2017, nearly double what theyre handling today.

One zettabyte is one billionterabytes. Even if its just abunch of 1s and 0s, thats a boatload of data, and itsgrowing with no end in sight. Wheres it coming from?And more importantly, where are the servers that will

handle it?Cisco divides all of that data center traffic into threeclasses: traffic that stays in the data center (76 per cent),traffic that flows between data centers (7 per cent), andtraffic that flows from the data center to consumers overthe internet or network (17 per cent). Heres a fact thathits home: Put all that data center traffic together, and in

2011, 70 per cent of it was handled in on-premises datacenters. By 2017, only 31 per cent will be handled at anorganizations physical site, leaving 69 per cent to behandled in the cloud.


16/88

8

There are quite a few factors driving this trend toward thecloud. Uptime Institutes 2012 Data Center Industry Study

reports that 27 per cent of companies are looking for costreductions, 23 per cent are motivated by an increase inscalability, 13 per cent cite pressure from customers orusers to improve access to data, and 13 per cent aretrying to build a more agile, responsive company arounda cloud-based data center.

Theres clearly some momentum here. Yet IT profes-

sionals are cautious by nature. Some look at cloud-baseddata center solutions and wonder about offsite security,deployment time, change management issues and otherunknowns. Their default mindset is, Were getting by why take a chance?

Seven Good Reasons to VirtualizeVirtualization is designed for this world of fast-growingdata and cloud-connected businesses. Today, a majorityof the worlds data centers have at least some degree ofvirtualization in place, and for good reasons: It makesthem more competitive and helps them manage theircloud-based workloads. Here are seven factors thattogether make a business case for virtualization.

Lower costs:To run a lot of servers, youll need tobuy a lot of power and remove a lot of heat. Add tothat the cost of buying, installing, upgrading andmaintaining servers and the costs can quickly spiralout of control. Businesses that go virtual areresponsible for less physical hardware and the

expenses that go with it.


17/88

9

Faster provisioning:Provisioning resources for atraditional data center can take some time and its

often a one-way street. Once resources areprovisioned, it can be difficult to reprovision them.Often its easier to just buy new servers, storage orother network assets. By comparison, virtualizedenvironments can be reprovisioned with a fewclicks. Does a workgroup need more storage? Doesthe lab need extra computing power for a few

weeks? Fire up your virtualization dashboard andits done.

Fewer business disruptions:When a physicalserver fails, it can take days or weeks to replace it.Someone on the IT team needs to buy and reinstallnew hardware. And if you dont have a current full-image backup of the server, youll also need toreinstall the operating systems and applications,configure directories, restore settings, reconfiguredatabases, and restore as much of the data as youcan find. In a virtual environment, failed servers arebypassed and data keeps flowing.

Easier backups: In a virtualized environment, its

easy to have full backups or snapshots of yourvirtual servers, virtual machines, databases, filesand settings. Virtual environments can be backed upfrom the system level to the object level as often asyou want, and its all transparent.

No vendor lock-in:Virtualized data centers have alevel of abstraction between hardware and software.That means a virtual machine doesnt know or carewhat server hardware it runs on. Youre not tied toone vendor for your hardware, operating system orcloud platform.


18/88

10

Greater efficiency: Data centers typically run 24/7so servers are plugged in and using energy around

the clock. However, according to a September 2013Gartner report,Maverick Research: Peer-to-Peer

Sharing of Excess IT Resources Puts Money in theBank, Other IT infrastructure utilization rates canbe quite low as well for example, servers oftenhave a utilization rate of less than 10%, while storageutilization is a bit higher but still below 50% in many

cases. Virtualization cuts this back with sharedhardware, software and infrastructure.

Head-start to the cloud: Virtualization is a strongstep in the direction of the cloud a step that mostcompanies are already taking. In fact, 55 per cent ofbusiness-unit aligned developers in IT operationsconsider private cloud a top infrastructure priority,

according toPredictions For 2014: Cloud Computing,a December 2013 report from Forrester Research,Inc. This is further proof that all the pieces of amodern data center are being influenced, at least inpart, by the cloud. Companies that virtualize theirservers and abstract their day-to-day operationsfrom the underlying hardware are positioned to

leverage cloud services if and when it make sense.

As with all change, there are growing pains as well assignificant opportunities. One thing is certain. In lookingat all the trends and statistics from industry watchers, itsclear that virtualization is here to stay, and thats a goodthing especially for those who help drive the evolution.


19/88

Chapter 2

The Evolution of Virtualization

In This Chapter

Hypervisor 101

Todays challenges

Tomorrows data center

T

he whole idea behind a virtualized data center is tohave a more efficient environment thats easier tomanage and orchestrate. Hypervisors got the ball rollingby providing a way for applications to share computeresources transparently.

Hypervisor 101: Sharing the LoveA hypervisor is software that is installed on a hostcomputer. That computer and the resources it controlsthen become part of a larger pool of resources that can beshared by virtual machines (VMs) on the network. Thehypervisor controls how those resources are accessed bythe VMs. There are multiple types of hypervisors:

Bare metal hypervisors: These are commonly calledType 1 hypervisors, and run directly on a physicalhost server. Type 1 hypervisors directly control theserver hardware and manage guest operating


20/88

12

systems. They partition the server into multiple VMsthat operate independently but share the same

network resources. Hosted hypervisors:Commonly called Type 2

hypervisors, these are installed as virtualworkstations on top of a servers existing operatingsystem, such as Windows Server, Linux or a customoperating system. The host OS has direct access tonetwork resources and hardware, and manages OS

connections to those resources. The hostedhypervisor coordinates calls between the VM andthe resources it needs to access, including CPU,memory, disk storage, and the network.

Todays ChallengesHypervisors brought the benefits of virtualization to thedata center. But with progress came challenges that arefamiliar to most IT managers. A Forrester Research paper,Expand Your Virtual Infrastructure with Confidence andControl, identifies the following common challenges intodays first-generation virtualized data centers:

VM sprawl: Creating new virtual machines is so easythat VM workloads tend to propagate quickly. Theresult is VM sprawl, where more and more VMs areadded to a virtualized infrastructure, adding timeand costs to manually monitor, manage and secure.

Over provisioning:Around 44 per cent of small andmidsized businesses surveyed by Forrester run sixor fewer VMs per physical server. That number hasthe potential to be much higher, probably 25 or moredepending on the workload and specifics of thehardware. But without capacity management tools,these businesses lack the visibility they need to


21/88

13

optimize the virtual environment. As a result, theyremissing out on the efficiencies and cost savings of

virtualization, and are limiting their return oninvestment.

Spotty visibility and monitoring:Most first-generation virtualized data centers lack the tools ordont have policies in place to monitor their virtualinfrastructure on a continuous basis. Virtualenvironments are monitored only when theres a

problem or a capacity shortfall. Worse, themonitoring tool of choice is often a spreadsheet,which makes it difficult to draw insights into whatsnot working, and why. Its a reactive rather thanproactive stance, and thats a hard way to run acompetitive business these days.

Complex troubleshooting: Two-thirds of todaysbusinesses need from half a day to a full week toresolve IT performance issues. Only one-third arecurrently able to resolve most problems within anhour. For IT to keep up with a modern business,complex troubleshooting needs to happen faster andunplanned downtime needs to be avoided.

The Next Generation:Virtualization 2.0The next generation of virtualization referred to asVirtualization 2.0 is making a big leap forward.

The evolution is similar in some ways to the dramatictransformation of mobile technology. (If you need areminder, type cell phone 1980 into your favorite searchengine and check out the images.) The first generation ofmobility solutions, though primitive by todays standards,


22/88

14

introduced mind-blowing new functionality the ability tomake a phone call without a landline. The second

generation, smart phones, upped the ante by literallyputting a computer in every pocket, forever transforminghow consumers use their phones.

The changes underway in virtualization are just asdramatic. Heres what the next-generation data center willlook like with Virtualization 2.0.

Visible: With integrated management capabilities,next-generation data centers will have the visibilitythey need to truly see every layer, platform, app anddevice in their virtual environment. Virtualizationwill extend beyond the server to include storage,backup, and the entire network ecosystem, so ITteams will be able to see the entire ecosystem

theyre responsible for managing. Optimized: Over-provisioned capacity will be

reclaimed and put to use, so available resources willsee higher utilization rates and data centers will seelower costs for hardware, maintenance, power andcooling. Advanced capacity management will allowIT to play a crucial role in helping the business plan

resources ahead of time and increase profits. Predictive: Data centers will proactively analyze

performance data, help you spot trends, and identifythe root cause of performance drift as it occurswithin the data center. Some fixes will happenautomatically, others will trigger smartalerts thatprovide actionable recommendations to fix problems

fast. Smart alerts are selective they learn thetypical behaviors of VMs, devices and users in yourvirtualized environment, and avoid sending falsealarms that flood inboxes and eat up nights and


23/88

15

weekends. Youll get fewer, more meaningfulnotifications based on policies that you define, and

can find and fix problems fast before they causedowntime.

Always on: Automated, agentless failover/backupwith built-in redundancies across a massive virtuallandscape will eliminate the threat of downtime. If abank of servers, an entire facility or even an entireregion goes down, the data center will be unaffected.

Customized: Custom dashboards, reports and viewswill allow stakeholders outside the data center to see and understand how the IT infrastructure isrunning, and how it can run even better. Virtuali-zation management tools will monitor the wholeenvironment, analyze workloads, detect anomaliesand deliver insights via a custom dashboard.


24/88

16


25/88

Chapter 3

Virtualization 2.0: Starting withOperations Management

In This Chapter Capacity management

Performance monitoring

W

hether youve already virtualized your datacenter or are just getting started, its clear thatthe next generation of virtualization solutions need toaddress the issues of VM sprawl, over- or under-provisioning of resources, lack of insight into the ITenvironment, and the amount of time spent finding andfixing issues. The answer with Virtualization 2.0 isintegrated operations management, which gives ITteams the power to improve application performance,reduce operating and capital expenses, and make surebusinesses have what they need to be productive andresponsive. Heres a high-level look at what integratedmanagement brings to the table for todays virtualizeddata centers.


26/88

18

Capacity ManagementOver-provisioning is expensive. Its common to seevirtual machines with far more capacity than they needto support peak demand. As a result, server resourcesoften sit idle, using electricity, generating heat, andadding time and cost for IT but contributing verylittle to the operation of the business.

Under-provisioning can be risky with exposure toperformance issues. If a mission-critical workloaddoesnt have the processing power, storage and othernetwork resources it needs, VM performance suffers,and so does the business.

Capacity management: What is it?Capacity management helps identify idle and over-provisioned VMs to reclaim excess capacity andincrease VM density without impacting performanceThe key is insight into the data center -- knowing howmuch resource each VM really requires, amount ofresources available and how much will be required inthe future.

Companies that see the entire landscape have anadvantage. They get historical as well as real-timeinsight into the IT environment, and can use operationsmanagement tools to spot underused resources andsee the source of a problem for example, idle,powered-off or super-sized VMs. Dashboards make iteasy to see the resources that are available and makeinformed decisions about provisioning. And withpredictive analytics, its easier to anticipate futureneeds and provision the right amount of resources forfuture workloads.


27/88

19

A recent Management Insights study found thateffective operations management can provide

additional benefits on top of the gains that virtualizeddata centers have already achieved, including:

34% Increase in capacity utilization

36% Increase in consolidation ratios

30% Increase in hardware savings

26% Decrease in diagnostic & problem resolutiontime

How its done: Millennium Pharmacyoptimized IT resources with VMwareLike most healthcare organizations, MillenniumPharmacy Systems faced increasing economic pres-

sures that were also felt by its client nursing care facili-ties. This drove Millennium to develop a lean IT serviceand infrastructure model that could better meet itsown needs and pass on cost savings to customers.

To achieve this, Millennium sought a vendor that coulddeliver a more cohesive virtual environment managedwith fewer resources and provide maximum insight

into data center health and utilization. After havingsuccessfully virtualized three server farms containing25 physical machines on VMware vSphere software,VMware was the natural choice for these next steps inthe companys virtualization journey.

After piloting VMware vSphere with OperationsManagementin a proof of concept deployment,

Millennium realized the value of upgrading to a virtual-ization platform that provided insight into workloadcapacity and health, including opportunities to


28/88

20

optimize systems to cut costs and reclaim overprovi-sioned resources.

[It] was almost jaw dropping in terms of thememory and resources we would get backby using this product, said Leon Ravenna,Millenniums Vice President of Infrastructureand Operations and Information SecurityOfficer. Some of the results were almostunbelievable in terms of what they recom-

mended. It drove us to be more efficientthan we had been, and it taught us how tolook at VMs differently than physicalhardware.

Optimized resourcesBy using vSphere with Operations Management capac-

ity planning tools, Millennium discovered that everyvirtual machine was oversized. It quickly improved theefficiency of the entire data center by resizing each VM.

Our entire data center is much more effi-cient, much more trimmed down and able toutilize more VMs on less hardware, saidMillenniums Lead System Engineer, Craig

Clark. Now 94 per cent virtualized,Millennium has reduced the footprint in itscustomer-facing data center that runs itsproprietary MPSRx medication fulfillmentsystem from 25 physical machines down tojust four.

Streamlined managementIn addition to optimizing resources, Millennium stream-lined its data center management by using the single-pane-of-glass view provided by vSphere withOperations Management.


29/88

21

Prior to vSphere with Operations Management,remembered Clark, we had three different vCenter

Servers that required me to constantly toggle amongthe three to see each environment. Now, Clark can seeand keep track of his data centers through a single, uni-fied console.

When I first start my day with vSphere withOperations Management, I go straight to thedashboard, which gives me a global view of

the health, risk, and efficiency of my entireIT environment, said Clark. From there, Imable to drill down to see how each compo-nent of my data center is working, includingthe clusters, hosts, VMs, and the data stores,as well as any faults that happen to beoccurring. I can then quickly double-click to

get to the individual fault and determineexactly what it is and what action I need totake.

Business Impact

Higher levels of reliability

Fast recognition and resolution of issues saves time andreduces end-user impact

25 per cent increase in efficiency allows fewer resourcesto deliver more value

Ability to capacity plan reduces capital expenditures andensures server optimization

Increased VM density leads to lower capital and opera-tional expenses


30/88

22

Performance MonitoringIT lacks the tools to aggregate and analyzeperformance data quickly enough to take proactiveaction and reduce downtime. Legacy monitoring toolsused by IT are based on low-level metrics and staticthresholds. This leads to a lack of understanding onroot cause, alert storms, and false positives.

It can be difficult in a virtualized data center in anydata center, for that matter to get to the root cause ofperformance problems. Sometimes performance simplydegrades slowly over time. And like the frog in theslowly heated pot of water, businesses may not noticea problem in the data center until its too late.

Performance problems might be tracked to a change in

virtualization policy, a network performance issue, amemory bottleneck, under-provisioning, power outages countless culprits come to mind. Good luck findingthe root cause in a hurry if you need to launch multiplemanagement consoles and want to compare whatyoure seeing across server, storage and networkenvironments. No wonder performance problems areso hard to fix.

Performance monitoring: What is it?The advanced performance monitoring solutions ofVirtualization 2.0 provide a world view of the ITenvironment on one dashboard. Bottlenecks andperformance issues can be traced to the source.

There are two key features of an effective solution: Predictive Analyticsanalyze server performance

data, establishes dynamic thresholds that adaptto the environment, and provides Smart Alerts


31/88

23

about health degradations and performance bot-tlenecks to drive proactive action and policy-

based automation Operations Consoledisplays key performance

indicators in easily identifiable colored badgesand provides a comprehensive view into what isdriving current and potential future performanceand capacity management issues in one place

Troubleshooting gets faster, easier and cheaper withless time wasted searching for issues. The dashboardmakes it easy to filter through VMs or server clusterswhere performance levels appear to be degrading, get ahandle on whats happening, and make informed deci-sions on how to proceed.

Did the problem start within the last six hours or has it

been a long, slow slide? Is there a bottleneck in thesystem? Is it related to the CPU, memory or I/O? Is theissue limited to a particular VM, its peers or the host?With the integrated management tools of Virtualization2.0, IT has all the answers.

How its done: Cornerstone optimized IT

performance with VMwareCornerstone Home Lending, Inc. required a reliable vir-tualization platform to support growth ambitions of upto USD $10 billion in annual loans in the next few years.It found the ideal solution to fit its business needs inVMware vSphere with Operations Management.

With the help of partner Computex, Inc., Cornerstonewas able to implement 45 hosts and close to 100 virtualmachines within a very short timeframe. Today, thesolution provides insight into past and present asset


32/88

24

health through a comprehensive, centralized console,and enables best-in-class platform-wide management.

Ricky Caldwell, Cornerstones Director of ServerOperations, Architecture and Infrastructure, sees thecapacity planning and performance monitoring offeredby vSphere with Operations Management as vital to hisorganization.

Im able to open up VMware vCenter and

with everything in a common console, it givesme one full view to my entire environment,Caldwell said. Im not required to open upfive different management consoles in orderto see my server, my storage, my network. Itsall in one location, which makes it very easy.

Maximizing system performance

With vSphere with Operations Management, we get avery in-depth view of whats happening on variouslevels, on a day-to-day basis, a weekly basis, or amonthly basis, added Cornerstones Server OperationsProject Lead, Stan Thampi. Alerts tell me right away if Ihave any emerging performance issues that need to belooked at, before they impact the business.

To maximize system performance, Cornerstone utilizesvSphere Distributed Resource Scheduler (DRS) pro-vided within the vSphere with Operations Managementplatform. With DRS, Cornerstone can load balanceresources to virtual machines by deploying new capac-ity to a cluster or automatically migrating virtualmachines at any time without any service disruptions.

Leveraging a holistic viewvSphere with Operations Management collects the met-rics from vCenter Server to provide a holistic view of


33/88

25

Cornerstones IT infrastructure. The comprehensivedashboard includes health, risk, and efficiency scores,

so managers quickly understand IT service levels andworkload capacities.

Since implementing vSphere with OperationsManagement, Cornerstone can now better identifycapacity shortfalls, stressed workloads, and over-provi-sioned servers. These insights allow the company tooptimize the use of its existing resources and make IT

investment decisions with certainty.

Reducing hardware costs The switch from physical to virtual servershas resulted in hardware cost savings closeto 70 per cent. Were no longer spendingfour or five thousand dollars per physical

server. We can now leverage the virtualhardware we have in place for a third of thecost, added Caldwell.

In addition, Cornerstone has reduced operatingexpenses since virtualized infrastructure require lesspower and cooling.

Business Impact

Comprehensive insight into IT

Improved resource planning to better inform IT invest-ments and decision making

70 per cent reduction in hardware costs

Ensured uptime with reliable virtualized infrastructure

Reduced time spent on identifying and resolving systemissues


34/88

26


35/88

Chapter 4

Virtualization 2.0: Deep Dive

In This Chapter

Virtualization 2.0 technologies beyond the server

Operations management for virtualizedenvironments

Virtualized backup

Virtualized storage

Virtualized networking

If weve learned anything about data center virtualizationover the past few years, its that cloud-based data

environments are a mosaic. Most businesses use a mix ofprivate and hybrid cloud services to meet the needs of a

bewildering number of devices, operating systems,applications, users and work groups.

Virtualization 2.0 fits this borderless world pretty well. Infact, borderless is a good description for next-generationvirtualization, which goes beyond servers and VMs toinclude operations management, backup, disaster recovery,storage, network operations and more.

It makes sense. A data center is the nerve center of abusiness. Virtualizing just part of it limits its benefits andreduces the return on investment. In this chapter, youllmeet the four horsemen of the virtualized data center:


36/88

28

operations management, backup, storage, and the network.Youll see how they work and how they work together to

create a modern, fully optimized IT environment thatsready for whatever a cloud-based business can throw at it.

Well reference specific VMware products from this pointforward, so you get both a micro and macro view ofVirtualization 2.0 and its role in a state-of-the-art data center.

Operations Management forVirtualized EnvironmentsIf your employees and customers are happy andproductive, thank IT ops. IT operations management playsa central role in how and how well knowledge workersdo their jobs.

Its no exaggeration to say that customers and employeesrely on IT applications and infrastructure every minute ofevery day. Nothing has a greater impact on their ability todo their jobs. Yet many data centers lack the monitoringtools they need to keep the data center humming. Many uselegacy monitoring tools in some cases, spreadsheets fortroubleshooting and provisioning.

Most legacy monitoring tools were not designedto keep up with a modern, virtualized environ-ment. They force businesses to be reactiveinstead of proactive: all too often, IT admins gethit with alert storms and false alarms orworse, no warning at all when the perfor-mance of the net-work or an application drifts

out of spec or begins to bottle-neck. For IT, itoften means fire drills and late nights.

Since most IT departments are stretched thin, they cantafford to spend more time, budget and resources manually


37/88

29

identifying and resolving performance and capacity issues.In many cases, a small IT team is supporting an organiza-

tion that has significant competitive pressures; siloed man-agement tools make their job harder than it needs to be.

The challenge: Improve efficiency, and fastHow efficiently can you deliver IT services, deploy ITresources, and enforce IT policies and compliance mea-sures? The answer affects your profitability. With modern

tools such as those found in the VMware vCenterOperations Management Suite, data centers can improveefficiency and achieve three key benefits:

Faster service delivery:With insight across the entirevirtual environment, including applications,infrastructure, devices and VMs, its easier to identifyperformance drift and fix performance, capacity and

configuration issues before they cause downtime

Higher operational efficiency: By optimizing howresources are used and managed, over-provisioning isavoided and unnecessary purchases are reduced. Andit goes beyond cost efficiency. IT teams are also ableto work more efficiently, reducing fire drills and latenights, and focusing their skills to help the businessbe more competitive.

Improved control and compliance:By automatingconfiguration and compliance policies and monitoringresults, the burden on IT is reduced and many of thecosts of non-compliance are avoided.

How it worksvCenter Operations Management Suite delivers intelli-gent operations from applications to storage managingthe health, risk, efficiency and compliance of a dynamicenvironment from a single console.


38/88

30

Self-learning analytics, dynamic thresholds, and the abilityto automatically correlate application and infrastructure

performance help IT teams to proactively identify potentialcapacity and performance issues. Dynamic thresholds auto-matically adapt to the virtual environment, providing fewerand more specific alerts regarding health degradations, bot-tlenecks and capacity shortfalls. Smart Alerts combine mul-tiple symptoms into a single alert that identifies theunderlying issue and provides clear recommendations for

remediation. Smart alerts, along with the softwares abilityto map application dependencies to the underlying infra-structure, make it easier to find and fix problems faster.

Operations teams can create custom workflows forcommon IT operations and processes, all of which can beautomated based on policies they control. Common actions such as defining specific capacity thresholds, alert types,

notifications and configuration settings can be prioritizedaround the needs of the business.

Storage analytics provide visibility into infrastructure topol-ogy, usage statistics and events across host bus adapters,fabric and arrays using standard protocols. Automatedcapacity optimization allows operations teams to reclaimover-provisioned capacity, increase resource utilization and

eliminate the need for scripts and spreadsheets.

The Suite automatically detects, enforces and remediatessecurity hardening guidelines, configuration standards andregulatory compliance requirements across the virtual envi-ronment. Change events can be visually correlated withperformance data across physical and virtual infrastruc-ture, operating systems and applications, providing visibil-ity into performance degradation caused by configurationchanges.

Custom dashboards, reports and views help operationsteams visualize key performance indicators (KPIs), provide


39/88

31

role-based access and enable better collaboration acrossinfrastructure, operations and applications teams.

Virtualized BackupProtecting data in a virtualized environment can be tricky.Data is distributed across a lot of devices, platforms andfacilities, making it hard to control. Whats more, newvirtual machines can be spun-up easily, often without the

backup administrator knowing that the workload needs tobe part of the backup routine. Its a recipe for trouble.

Virtual environments that use conventional software toback up their virtual machines are adding complexity to anenvironment thats already complicated enough.Conventional backup software must be licensed, installedand maintained on every virtual machine, so that software

competes for virtual machines resources and impactsserver performance.

The goal: Simplify, simplifyIn a virtual world, new virtual machines need to bedetected as soon as theyre created and they need to beprotected when they move. Setup routines need to be fast

and easy; if theres a steep learning curve or if a backupsolution is overly complicated, chances are good thatnewer workloads and devices will go unprotected until theIT team can find time to address the gap.

Virtualized backup solutions are designed to integrate witha virtualized data center, providing backup scheduling andmanagement from a unified console. This improves control

over all assets virtual and physical while optimizing howstorage is consumed during virtual machine backups,reducing backup costs.


40/88

32

VMware vSphere Data Protection Advanced provides agood example of how a fully virtualized backup strategy can

simplify life for IT admins, and make sure the virtualizedenvironment is ready for fast disaster recovery. With virtu-alized backup integrated into the virtualized data center,IT can:

Manage backup operations across all virtual machinesfrom a unified UI

Perform agentless virtual machine backup andrecovery

Back up application data on virtual and physicalservers

Replicate virtual machine and application backups fordisaster recovery

Reduce backup storage footprint through advanceddata deduplication

Recover from data disasters quickly with ChangedBlock Tracking technology

Meet infrastructure SLAs

Reduce total cost of ownership through simple per-

CPU licensing and support

How it worksVMware vSphere Data Protection Advanced is a disk-based backup solution, deployed as a pre-built, Linux-basedvirtual appliance. During a backup, the software creates afull-image snapshot of the virtual machine. Data deduplica-

tion is performed with every backup operation to optimizestorage space. The vSphere Hypervisor web UI is used toselect, schedule, configure and manage backups. Specificvirtual machines or virtual machine disk (.vmdk) files can


41/88


42/88

34

Virtualized StorageVirtual machines require storage. Lots of it. The worldsstored data doubles nearly every 24 months. This createsconstant pressure in the data center where, all too often,legacy storage solutions pre-date the shift to virtual infra-structure. It creates a functional gap and no small amountof stress since the needs of the business are being met inmany areas, but rarely in the area of storage.

It used to be easier. In the days before VMs, traditional stor-age solutions such as storage-area networks (SANs) pro-vided a centralized approach to storage, so multiplemachines, workgroups or departments could share storageresources. VMs added a layer of complexity to the storagepicture. IT managers suddenly were expected to providethe same level of service to a virtual environment that was

available with a SAN.

The answer: Virtual storage-area networksA virtual storage-area network is created by software thatcan pool storage components in standard x86 servers andpresent them as a shared datastore through the hypervisorto virtual machines. VMware Virtual SAN pools the solid-

state drives (SSDs) and hard-disk drives (HDDs) that areinside physical servers, and make them appear as a single,shared storage resource. Theyre designed for todays virtu-alized, software-defined data centers, removing the com-plexity of deploying and managing storage for virtualenvironments. That means IT managers can set and changeapplication service levels on the fly, in software, and control

storage thats as fluid and dynamic as the virtual datacenter it serves.


43/88

35

Modern tools, like Virtual SAN, address three big require-ments for virtualized data centers:

Simpler storage:Policy-driven storage simplifies andstreamlines provisioning and management for soft-ware-defined data centers. VM-level storage policiesautomatically match requirements with underlyingstorage resources. Many manual storage tasks areautomated, improving efficiency while reducing costs.

Higher performance:When read/write caching usesserver-side flash, storage commands are executedfast. A virtual SAN solution actually delivers betterperformance than most virtual appliances or externalarrays.

Lower cost: When storage can be scaled as needed,a storage environment can grow incrementally, in

small batches. No more over-provisioning or buyingforward. Leveraging server-side economics allows youto enjoy savings from reduced markups on all storagecomponents. Hardware-agnostic design means youcan leverage your existing infrastructure, and havethe flexibility to design the solution for a wide rangeof performance and capacity requirements. In addi-

tion, UI familiarity means a short learning curve sooperational productivity is improved, saving timeand money.

How it worksVirtual SAN creates a shared datastore for virtualmachines by pooling the SSDs and HDDs inside servers and

making the resources available to VMs. Once a Virtual SANis set up, storage can be provisioned to new VMs and exist-ing VMs can be moved from another datastore into theVirtual SAN. The Virtual SAN interface is integrated into a


44/88

36

console with the vSphere hypervisor web client, allowing itto work seamlessly with virtualized applications for

backup, data replication, cloning, snapshotting and more.Different storage policies can be identified for each VM akey consideration for virtual environments. This allows ITadministrators to do things like:

Specify the number of hardware failures a VM cantolerate.

Dedicate a specific percentage of flash storage to a VM. Identify the number of hard disks that should be

provisioned for a VMs objects.

Specify the overall percentage of capacity that shouldbe reserved for each VM.

Change policies on the fly from the vSphere web client.

Virtualized NetworkingServer and storage virtualization solutions have trans-formed the data center. Network virtualization is doing thesame. The automation, the cost savings, the benefits of con-solidation and on-the-fly provisioning and software-based

control all are coming to a network near you. Good thing,too, because without network virtualization, the full poten-tial of data center virtualization is not being realized.

Traditional networks and network services simply cantkeep up with the rapid-fire pace of a virtualized data center.Manual provisioning, vendor-specific hardware and rigidnetwork topologies all affect how traditional networks oper-

ate and how the IT team spends its time.On a daily basis, IT teams with traditional networks strugglewith time consuming and error prone manual provisioning,


45/88

37

which is not only slow but also error prone. Since work-loads are restricted by physical topologies, network opera-

tors spend a lot of time performing manual box-by-boxconfiguration of VLANs, ACLs, firewalls and more. IT admin-istrators are often forced to over-provision the network dueto VM sprawl. Add over-subscription, security blind spotsand siloed management and its easy to see why most ITteams have their hands full.

The mission: Make life easierVirtualized networks are designed to address these issuesby aligning with the underlying design philosophy of avirtualized data center. The VMware NSX networkvirtualization platform shows how the software-defineddata center approach turns into actual advantages for ITteams. For example, a fully virtualized network means:

Streamlined operations:Network engineersdramatically reduce or even eliminate manualnetwork and security configuration changes. They cannow programmatically create, provision, snapshot,move, delete and restore complex networkenvironments all in software.

Greater flexibility: Traditional networks are rigid, andtheir functionality is slow to evolve. Networkvirtualization transforms the underlying physicalnetwork into a pool of network capacity that can beprovisioned and repurposed on demand.

Unrestricted workload mobility and placement:Workloads can be placed or moved freely through the

physical data center environment while remainingconnected to any virtual network. This ability allowsany available compute capacity to be utilized at anytime, dramatically reducing the need to over-provision


46/88

38

servers to compensate for physical topologyrestrictions.

Bulletproof security: Rather than being tied to IPaddresses, network policies can be based on virtualcontainers, applications and Active Directoryidentities. Security policies are automatically attachedto VMs as theyre created, policies move with VMsand are deleted when VMs are deleted, all with nohuman intervention.

Disaster-proof operation:A virtual network makes iteasy to get a full-image backup of the applicationarchitecture and have it available for push-buttonrecovery in minutes, on any replacement hardware,with no compromise in functionality.

Lower cost: Network provisioning and configuration

are automated. Administration, monitoring andtroubleshooting are faster and easier. Networkvisibility and control spans both physical and virtualinfrastructure. Hardware independence allowsoperators to build their physical infrastructure withthe best price/performance solution from any vendor.

How it worksVMware NSX is based on a distributed system controllerwith traditional hypervisor and vSwitch. The entire virtualnetwork is created in software, and is independent of theexisting physical infrastructure, so NSX works with anyapplication, hypervisor, network infrastructure or cloudservice, and can be deployed on any IP network, including

current or next-generation fabric architectures from anyvendor.


47/88


48/88

40


49/88

Chapter 5

Real-world Outcomesof Virtualization 2.0

In This Chapter Data center virtualization and standardization

Streamlined and automated data center operations

Security controls native to infrastructure

High availability and resilient infrastructure

App and infrastructure delivery automation

IT departments are responsible for solving some prettybig issues, and it goes beyond technology. Is it secure? Is

it in compliance? Can it recover from a disaster? How fast?How well?Does every department, team and employeehave whats needed to help the company compete and beprofitable?

IT has a hand in all of it.

Clearly, IT teams focus not just on the firedrill du jourbutalso on big-picture, macro outcomes that make their com-

pany more successful. Here are the top-five IT outcomesthat most data centers strive for. Virtualization 2.0 is thefoundation that makes each one possible...


50/88

42

IT Outcome: Data Center

Virtualization and StandardizationComplexity is a productivity killer. Overly complex datacenters bleed time and energy from IT teams, and oftenresult in unnecessary capital expenses, error-prone manualprocesses, disorganized topologies, siloed infrastructureand general lack of interoperability across the data environ-

ment. The end result is unplanned downtime the ultimatecomplication for any competitive business.

Virtualization 2.0 helps by simplifying the data center. Theentire environment compute, storage, backup, network is created and managed in software, so the underlying com-plexities are masked. These software-defined data centersimprove efficiency through hardware consolidation a bot-

tom-line priority for most companies. With fewer machinesto buy, configure, install, maintain, troubleshoot and sup-port, IT can help the business run leaner and moreprofitably.

Data center consolidation and standardization can beaddressed in three stages of maturity with Virtualization 2.0solutions:

Stage 1: Compute virtualizationGet more mileage from your existing assets and the savingsadd up. This is the starting point for most businesses thatare looking to reduce costs.

Required capabilities (what you need to be able to do):

Compute virtualization. Business-critical applicationvirtualization. Big Data app support.

Business outcome (how your business benefits):Dramatic CapEx and OpEx reduction result in a datacenter with better-than-public-cloud unit costs.


51/88

43

Stage 2: Network virtualization and

software-defined storageOnce the hardware environment is simplified, software andnetworking are next. Streamline here for additional savings.

Required capabilities:Run-book automation to sup-port data center migrations, software-defined storage,and network virtualization

Business outcomes:Extend CapEx and OpEx reduc-tions across the entire data center infrastructure tomaximize the return on your investments

Stage 3: Hybrid cloudBig efficiencies can be gained by selectively deploying mul-tiple aspects of a business in the cloud.

Required capabilities:Ability to extend compute, stor-age, network, backup and disaster recovery resourcesto any cloud platform.

Business outcomes:Expand capacity and servicewhen you need them most without having to invest innew infrastructure.

IT Outcome: Streamlined andAutomated Data Center OperationsService-level agreements need to be met. To get there, datacenters either over-provision the network so theyre readyfor peak flows, or under-provision the network and risk slug-

gish application performance, capacity shortfalls and lack ofavailability when its needed most. The end result is a busi-ness disruption thats unplanned, unwelcome andunnecessary.


52/88


53/88

45

Required capabilities:Support third-party manage-ment packs for Microsoft, SAP and more. Provide role-

based access to customizable dashboards. Business outcomes: Faster diagnosis and trouble-

shooting. Fewer alerts, reducing fire drills for IT.Shorter mean-time-to-identification, improving servicedelivery to better meet service-level agreements.

Stage 3: App health monitoring, quality-of-

service managementAre your business-critical apps running as well as theyshould? Managing this part of your business can bring seri-ous gains.

Required capabilities:Manage business-critical appli-cations from any vendor (Microsoft, Oracle, SAP).

Enforce regulatory compliance standards. Provide sin-gle-console visibility across applications, storage andnetwork devices.

Business outcomes:Dramatic improvements in ser-vice quality. Fewer manual errors in configuration.Improved uptime on tier 1 applications.

IT Outcome: Security ControlsNative to Infrastructure

Few things keep an IT administrator up at night like securityand compliance concerns. With workloads, files and usersspread far and wide across the virtualized data center, its

hard to be confident that every single data door is locked,that every single compliance regulation is being met forevery user on every device.


54/88

46

Virtualization 2.0 helps by enhancing the software-definednetwork with policy-based security and compliance tech-

nologies, thus allowing workloads to be configured andmonitored according to precise rules. These control mea-sures remain in place regardless of where workloads aredeployed in a private or hybrid cloud.

Security and compliance can be addressed in three stagesof maturity with Virtualization 2.0 solutions:

Stage 1: Network securitytied to workloadsSecurity and compliance policies need to be automatedacross the network. Every virtualized data center needs toaddress this baseline functionality.

Required capabilities:Layer 2 to layer 7 network virtu-

alization and security. Security and network servicesfollow workloads. Ability to introduce network serviceson the host (VM) level. Hardware-independent networkand security services. Automated networkinfrastructure.

Business outcomes:Faster deployment of network andsecurity resources. Increased hardware efficiency

through multi-tenancy. Next-generation network secu-rity. Increased VM security.

Stage 2: Advanced 3rd partysecurity services integratedinto network platform

As a virtualized network matures, additional refinementshelp address compliance and monitoring across the virtualinfrastructure.


55/88

47

Required capabilities:Integration with 3rd party secu-rity service providers, providing multi-level security to

the infrastructure, including advanced firewall services,intrusion prevention systems and network antiviruscapabilities, personalized and configured via policieson the VM level.

Business outcomes:Better multi-layer security withlow administration overhead.

Stage 3: Extension of network andsecurity services to hybrid cloudThe final stage of network maturity brings the network andall security and compliance services into the cloud.

Required capabilities:Streamline and secure theextension of network to hybrid clouds. Move infra-

structure to any cloud platform without having tochange network and security configuration.

Business outcomes:Easier workload migration acrossthe virtualized environment, with security baked in.

IT Outcome: High Availability

and Resilient InfrastructureStuff happens. Power grids go down, storms hit, virusesspread. The threats never stop. Nor does the need for net-work access. Data centers are expected to operate 24x7, befully secure, protect sensitive data and meet every compli-ance requirement. After a disaster passes and theres still

an inch of water in the conference room, workers expect tobe back online with no interruption, with every workload,app, file and setting right where they left it.


56/88

48

Thats the expectation for a modern data center.Virtualization 2.0 can help by simplifying how routine back-

ups and disaster recovery processes are managed. Withhigher levels of automation, non-disruptive testing andhigh-performance data protection, Virtualization 2.0 pro-vides expanding degrees of protection to cover VMs, clus-ters and the entire data center. Operations managementtools, snapshots, live migration tools and other capabilitieshelp reduce or even avoid business impacts before they

cause a disruption. The net effect is less risk of businessdowntime and continuous, seamless data center operationacross private and hybrid clouds.

Business continuity can be addressed in three stages ofmaturity with Virtualization 2.0 solutions:

Stage 1: High availability, data protection,

and disaster recovery-as-a-serviceWheres your next data disaster coming from? Even if youhad a crystal ball and could see the future, you would needto address these baseline items.

Required capabilities: Back up hundreds of VMs pervirtual appliance, Provide agentless, image-level back-

ups to disk for any virtualized app. Conduct datadeduplication with each backup. Use across sites andto hybrid cloud.

Business outcomes:Zero service disruption whendisaster strikes. Guaranteed resource availabilityacross the business, so outages are avoided. Full dataprotection for all workloads for fast recovery from

data disasters.


57/88

49

Stage 2: Automated disaster recovery

for vSphere environmentsWith automated workload recovery across sites, the disas-ter recovery solution brings cost-effective data protectionand reduced downtime for apps and services.

Required capabilities:Non-disruptive DR testing forvirtualized environments. Recovery setup in minutes.Support for all applications. One-click failoverinitiation.

Business outcomes:Comprehensive, low-cost disasterrecovery with failback prior settings and data areresynchronized to pre-disaster locations to greatlyreduce business disruptions.

Stage 3: Self-service, policy-based DRprotection for apps

Advanced DR policies for failback create an ultra-resilientdata environment one that can recover quickly fromalmost any disaster.

Required capabilities: Define and provision DR pro-tection tiers to virtual machines. Self-service, policy-based provisioning of predefined DR tiers.

Business outcomes:Reduced admin time to provisionDR-protected workloads. Quicker time to market forDR-protected workloads.


58/88


59/88

51

Stage 2: Application and middleware

delivery automationAutomating how middleware and apps are provisioned canfurther accelerate virtualized operations, especially assome services move to the cloud.

Required capabilities:Automated middleware andapps provisioning. Policy-based network and storageservices. Service costing.

Business outcomes:Improved agility. Less time toprovision and deploy complex applications. Easy toincrease capacity on demand extend workloads tothe hybrid cloud as needed.

Stage 3: Heterogeneous and hybrid cloud

Fully optimized environments transparently support anyworkloads across a combination of hypervisors and hard-ware in private and hybrid clouds.

Required capabilities: Application release automa-tion. XaaS. Provisioning and migration of workloadsfor multi-hardware and multi-hypervisor across pri-vate and hybrid clouds including vCloud Air, AWS,

and those based on OpenStack. Business outcomes:Less IT time and cost to provision

fully loaded VMs. Lower CapEx thanks to optimized,consolidated storage and networking infrastructure.


60/88

52


61/88


62/88

54

Assemble a detailed plan that outlines the full scope ofthe project and its phases. Work with your finance

team members to determine total cost of ownership(TCO) and your projected return on investment (ROI).If you need new hardware such as servers, storagearrays or networking gear, put it in the budget.

Remember, the cost of new hardware mightbe offset by savings in other areas, such as

maintenance or operating expenses.Next, decide which workloads will be your highest pri-ority for the physical-to-virtual (P2V) migration.For example, you might start with test and develop-ment workloads, then virtualize your Tier 2 applica-tions, and finally virtualize your Tier 1, mission-criticalapplications.

Before you roll out the new virtual environment, allowtime to test it thoroughly.

Record baseline performance on your currentservers and applications. Its important to havethis data before migration begins so you canbenchmark VM performance gains against native

performance levels. This helps justify the projectfor management buy-in.

Make sure your infrastructure is ready to providethe high levels of service your business demands.Build redundancy into the physical plant (espe-cially power and cooling), build redundant net-work and storage fabrics, and use quality gear

throughout. The business is betting on IT to getthis done right, so build it right and build it fromthe bottom up.


63/88

55

As always with such efforts, be sure to carefullyschedule the migration and expected down-

time for the workloads youre migrating. Usinga proven P2V conversion tool such asVMware vCenter Converter can help youminimize downtime and maximize automationduring your migration process.

vCenter converter converts both Windows and Linux

physical machines to a VMware virtual machine formatand brings them into your VMware environment, auto-mating the migration process semi-transparently foryour users.

Using Traditional vs. Virtual StorageShared storage improves availability and allows hyper-visors to leverage capabilities, such as VMware vMo-tion, to migrate running VMs across hosts forzero-downtime maintenance. Today, there are multipleoptions for shared storage:

Traditional external SAN or NAS array:Compared to virtual storage, a storage-area net-

work (SAN) or network-attached storage (NAS)solution can be more expensive and require moretechnical expertise because they need specializedhardware and IT staff. For organizations withavailable capital and larger IT environments, tradi-tional arrays provide deduplication, array-basedreplication and unified storage offerings (forexample, NFS, iSCSI, Fibre Channel).

Virtual storage: This option is simpler than SAN andNAS because you dont need to purchase, configureor maintain an external hardware array. For


64/88

56

businesses that need shared storage but do notneed all the features of an enterprise storage solu-

tion, a solution such as VMware Virtual SAN cansave capital expenses and ongoing managementcosts.

Sizing and Managing Shared StorageVirtualization allows you to pool your storage infrastruc-

ture, which gives you flexibility for optimal workloadplacement. You can place highly volatile I/O workloadssuch as Tier 1, mission-critical database applications onTier 1 backend storage, such as high-speed SSDs orenterprise-grade SAS disks. At the same time, you canmove test and development environments or rarelyaccessed data to slower and lower-cost storage to

reduce expenses over the long run.When sizing and managing your shared storage, youshould:

Monitor how much space is used on your existingphysical volumes, and also the number of IOPS (I/Os per second) your workloads use. This informa-tion can help you choose the right type and size

of disks for your new environment.

Calculate your storage needs, in both raw capac-ity and IOPS, on current and future workloads.Whats the best way to meet those needs? Do youneed the array-based replication or extremeamounts of capacity that a traditional storagearray can provide? Or could your needs be met bya more cost-effective Virtual SAN solution thatallows you to scale storage capacity and perfor-mance as you add physical host servers?


65/88

57

Take advantage of the storage efficiencies of virtu-alization. For example, on a traditional physical

server, adding or reconfiguring disk drives is diffi-cult, time consuming, constrained by availabledrive bays and can sometimes result in downtimeor data loss. In a virtual environment, physicalstorage devices are abstracted seperated fromthe virtual machine, so storage capacity can beadded without affecting the VM in any way.

Virtual disks, by the same token, can be easilyexpanded without requiring complex reconfigura-tion of physical storage devices.

Choose thin or thick provisioning of virtual disksfor individual VMs. Thick provisioning allocatesall the space for a virtual disk the moment youcreate it; thin provisioning allocates space as nec-

essary throughout the virtual disks life. If youhave a dedicated storage solution from a thirdparty, there may also be thin provisioning availa-ble at the array hardware level as well.

Using VMware vSphere with OperationsManagement, you can configure Storage I/Ocontrol to guarantee a certain amount of I/O

resources for each virtual disk, or enable StorageI/O Control to provide equitable access to storageresources for all VMs. This ensures that no partic-ular workload will dominate the resources of anyphysical array.


66/88

58

Addressing Security and ComplianceJust like physical servers, VMs need to have appropri-ate security and compliance policies in place.Remember:

If your business must comply with any govern-ment regulations, consider any audit rules thatapply. For example, will it be acceptable for eachworkload to share physical networks or virtualswitches? Must the data itself be on separatephysical storage?

As you set up policies and provisioning, keep inmind the challenge of managing sensitive datafrom different applications. Do the rules allow thatdata to reside with the data from other applica-

tions at the compute, networking and storagelayers?

Make sure you have a working management net-work with all management interfaces of physicalhosts, switches and other data center infrastruc-ture in the environment. Isolated management net-works provide higher security while preventing

VM traffic from interfering with managementtraffic.

You need to balance VM protection with perfor-mance by scheduling security scans and otherchecks for off hours. Also, consider using featuresbuilt into vSphere with Operations Management,such as VMware vShield Endpoint, which

offloads anti-virus and anti-malware processes.This approach can also lower the risk of antivirusstorms by centralizing those functions.


67/88

59

Using Operations Management to

Meet Business ObjectivesServer virtualization allows physical resources to beshared among many virtual servers, improving systemconsolidation. Make sure your mission-critical applica-tions have the resources they need to perform wellwhile ensuring they meet your companys compliance

and security policies. Things to consider: Define affinity rules for your VMs. For example, you

can define host affinity rules to keep VMs together, soa web server VM and its associated app and databaseVM are kept on the same physical server for high-speed virtual network connectivity. You can alsodefine anti-affinity rules. For example, you can keep

multiple database servers on separate hosts so if aphysical host fails, other database VMs will keeprunning.

Determine whether your applications must reside onspecific hardware for compliance or process reasons.

Make sure you determine the Recovery Time

Objective (RTO) and Recovery Point Objective (RPO)for each workload, so when youre creating your busi-ness continuity and disaster recovery plans, yourbackup and recovery policies are aligned with yourbusiness priorities.

Adding Operations ManagementAs the saying goes, those who ignore history aredoomed to repeat it. By monitoring performanceissues, resource shortfalls and other historical data onyour VMs, you can anticipate future spikes in memory


68/88

60

and CPU usage, and plan accordingly so critical appli-cations do not hit capacity limits. The tools of

Virtualization 2.0 make it easy to monitor and analyzeworkloads and diagnose problems, so you can keepyour business-critical applications and VMs operatingat peak performance.

Before you deploy, here are ten things you needto know. These tips will help you get the mostfrom vSphere with Operations Management,

our core virtualization product, vSphere, withthe addition of performance monitoring andcapacity management capabilities through thevCenter Operations Management Suite. Thiscombination creates the operations manage-ment platform for Virtualization 2.0.

Prepare Carefully to OptimizeResource Utilization

Early planning will help improve consolida-tion and ROI down the road. So before youinstall, find the guardrails with vCenter

Operations Manager. How many hosts, VMs,storage systems and clusters will you bemonitoring? Youll need this baseline infor-mation before you start.

Consider the Appropriate Permissionsfor Different User TypesWith vSphere with Operations Management, you canconfigure permissions and security, and then assignprivileges so authorized users have access to the rightassets in the management console.


69/88

61

For example, admins should be able to toucheverything. Help Desk staff should have the

permissions they need to fix day-to-dayproblems without inadvertently changingpolicies or settings. Consider the best way toset up multiple user permission levels inorder to maintain the security of the environ-ment while giving all staff members the abil-ity to fix day-to-day problems as they arise,

so your business keeps moving.

Model Your Virtual Environmenton Your Business StructureStructure your operations views with intelligent groupsthat make sense for your business. Group objectsbased on specific business needs, departments, loca-tions, and more to create a simplified view of your envi-ronment from the vCenter Operations dashboard.Tailoring groups to your specific business needs helpssimplify IT tasks, so IT staff is better equipped tomanage more systems, lowering administrator over-head and freeing up staff for innovation elsewhere.

Create Policies for EfficientResource ManagementUse vCenter Operations Manager to assign policies tocertain groups of resources, geographic locations, or

business units to customize badge thresholds (see nextstep), alerts and capacity management settings. Takeadvantage of the out-of-the-box policies included withvSphere with Operations Management that will meetmost of your business needs (e.g., production or test


70/88

62

environments, batch or interactive workloads) orcreate your own personalized policies.

Configure Badges for Quick Health ChecksBadges are a visual representation of infrastructure assets,providing scores and metrics that show the health, riskand efficiency of the virtual environment and monitoredobjects. Badge thresholds should be customized based on

your environments size and unique requirements. Badgesare assigned a numerical value ranging from 0 to 100 andare color-coded depending on the thresholds defined. Its asingle, integrated, real-time status indicator for all virtualinfrastructure resources, allowing IT admins to more easilymonitor the virtual environment.

Identify the Needs of Workgroupsto Configure Capacity SettingsEvery workgroup has different needs at different times. Aproduction team working on a product launch might needto be over-provisioned for a few months with extra CPUand storage. A development and test environment might

be fine with high-density, over-committed VMs andresources. With accurate capacity analysis, you canaccount for varying business needs and tap your massivepool of resources so every workgroup has what it needs.

Choose How You Want to be AlertedSmart alerts let you choose how you want to be notifiedby your management platform when a problem is devel-oping. vCenter Operations Manager learns typical behav-ior in your environment, so it provides fewer, more


71/88

63

meaningful alerts that let you know when there really is aproblem for example, when a threshold is exceeded or

an anomaly is detected. Similar to capacity settings, alertsare configured based on policies that you define. Alertsalso provide actionable recommendations so you can findand fix problems fast, before they cause downtime.

Set up Email Notifications

for Administrative AlertsTo monitor data center health and capacity from any-where, configure an optional SMTP server to activate emailservice for notification messages when problems occur.

You can set email notifications for all types ofalerts, so you can address problems as they

happen in real time, minimizing downtime. Foradministrators, its especially important to setup email notifications for administrative alertsto ensure vCenter Operations Manager isalways running properly.

Schedule Reports to Help AddressBottlenecks Before They OccurUse reports in vCenter Operations Manager to monitorcapacity and performance in the vSphere environmentand to help avoid bottlenecks. Its a good idea to sched-ule reports for regular intervals weekly, monthly,quarterly, whatever makes sense for your business.

You can also pull reports on demand for a real-timesnapshot of the IT environment, and use historicalrepo

VIrtualization 2.0 for Dummies

Documents

Transcript of VIrtualization 2.0 for Dummies