Virtualization for Dummies

Bernard GoldenMargaret LewisTim Mueting

Compliments of

• Create a dynamic data center

• Allocate memory where it’s needed

• Save energy, time, and money

• Improve security and scalability

Learn to:

Virtualization

2nd AMD Special EditionMaking Everything Easier!™

Open the book and find:

• A look at microprocessor advancements that help improve virtualization performance

• An overview of the different types of hypervisors

• How virtualization helps reduce power consumption in the data center

• An explanation of how memory is virtualized

Bernard Golden is a cloud computing expert and the author of Virtualization For Dummies. Margaret Lewis is currently director of Software Solutions for AMD, focusing on identifying the next gen software solutions for AMD products. Tim Mueting is responsible for AMD’s virtualization solution strategy.

ISBN: 978-0-470-64412-6 AMD tracking number: 42139-CNot for resale

Go to Dummies.com®

for videos, step-by-step photos, how-to articles, or to shop!

It seems like everywhere you go these days, someone is talking about virtualization. Technical magazines trumpet the technology on their covers. Virtualization sessions are featured prominently at technology conferences. Basically, virtualization is a technical innovation designed to increase the level of system abstraction and enable IT users to harness ever-increasing levels of computer performance.

• Virtualization’s background — understand where it came from and better know its moving parts

• Putting virtualization to work — what it can do for you and your company

• Hardware-assisted virtualization — AMD hard-ware features that help you get the most out of your virtualization environment

• Vir tualization for the desktop and client — server hosted desktops help improve manageability, security, and reduce costs

Understand how virtualization works and figure out whether it’s right for you

About AMDAdvanced Micro Devices (AMD) is an innovative technology company dedicated to collaborating with customers and technology partners to ignite the next generation of computing and graphics solutions at work, home, and play. AMD Opteron™ processors offer uncompromised design and deliver unprecedented performance-per-watt capabilities, providing the scalability needed to drive demanding and data-intensive applications.

These materials are the copyright of Wiley Publishing, Inc. and any dissemination, distribution, or unauthorized use is strictly prohibited.

VirtualizationFOR

DUMmIES‰

2ND AMD SPECIAL EDITION

by Bernard Golden, Margaret Lewis, and Tim Mueting

01_644126-ffirs.indd i01_644126-ffirs.indd i 5/3/10 11:48 PM5/3/10 11:48 PM


Virtualization For Dummies,® 2nd AMD Special Edition

Published byWiley Publishing, Inc.111 River StreetHoboken, NJ 07030-5774www.wiley.com

Copyright © 2010 by Wiley Publishing, Inc., Indianapolis, Indiana

Published by Wiley Publishing, Inc., Indianapolis, Indiana

No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except as permitted under Sections 107 or 108 of the 1976 United States Copyright Act, without the prior written permission of the Publisher. Requests to the Publisher for permission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permissions.

Trademarks: Wiley, the Wiley Publishing logo, For Dummies, the Dummies Man logo, A Reference for the Rest of Us!, The Dummies Way, Dummies.com, Making Everything Easier, and related trade dress are trademarks or registered trademarks of John Wiley & Sons, Inc. and/or its affiliates in the United States and other countries, and may not be used without written permission. AMD, the AMD Arrow logo, ATI, the ATI Logo, AMD Opteron, AMD-P, AMD Virtualization, AMD-V, CoolCore, FirePro, and combinations thereof are trademarks of Advanced Micro Devices, Inc. HyperTransport is a licensed trademark of the HyperTransport Technology Consortium. All other trademarks are the property of their respective owners. Wiley Publishing, Inc., is not associated with any product or vendor mentioned in this book.

LIMIT OF LIABILITY/DISCLAIMER OF WARRANTY: THE PUBLISHER AND THE AUTHOR MAKE NO REPRESENTATIONS OR WARRANTIES WITH RESPECT TO THE ACCURACY OR COMPLETE-NESS OF THE CONTENTS OF THIS WORK AND SPECIFICALLY DISCLAIM ALL WARRANTIES, INCLUDING WITHOUT LIMITATION WARRANTIES OF FITNESS FOR A PARTICULAR PURPOSE. NO WARRANTY MAY BE CREATED OR EXTENDED BY SALES OR PROMOTIONAL MATERIALS. THE ADVICE AND STRATEGIES CONTAINED HEREIN MAY NOT BE SUITABLE FOR EVERY SITU-ATION. THIS WORK IS SOLD WITH THE UNDERSTANDING THAT THE PUBLISHER IS NOT ENGAGED IN RENDERING LEGAL, ACCOUNTING, OR OTHER PROFESSIONAL SERVICES. IF PRO-FESSIONAL ASSISTANCE IS REQUIRED, THE SERVICES OF A COMPETENT PROFESSIONAL PERSON SHOULD BE SOUGHT. NEITHER THE PUBLISHER NOR THE AUTHOR SHALL BE LIABLE FOR DAMAGES ARISING HEREFROM. THE FACT THAT AN ORGANIZATION OR WEBSITE IS REFERRED TO IN THIS WORK AS A CITATION AND/OR A POTENTIAL SOURCE OF FURTHER INFORMATION DOES NOT MEAN THAT THE AUTHOR OR THE PUBLISHER ENDORSES THE INFORMATION THE ORGANIZATION OR WEBSITE MAY PROVIDE OR RECOMMENDATIONS IT MAY MAKE. FURTHER, READERS SHOULD BE AWARE THAT INTERNET WEBSITES LISTED IN THIS WORK MAY HAVE CHANGED OR DISAPPEARED BETWEEN WHEN THIS WORK WAS WRIT-TEN AND WHEN IT IS READ.

ISBN: 978-0-470-64412-6

Manufactured in the United States of America

10 9 8 7 6 5 4 3 2 1

01_644126-ffirs.indd ii01_644126-ffirs.indd ii 5/3/10 11:48 PM5/3/10 11:48 PM


http://www.wiley.com

http://www.wiley.com/go/permissions

Table of Contents

Introduction ....................................................... 1About This Book ........................................................................ 2Foolish Assumptions ................................................................. 2Icons Used in This Book ............................................................ 2

Chapter 1: The Roots of Virtualization . . . . . . . . . . . . . . . .3Why Virtualization ..................................................................... 3Coming Up with the Concept.................................................... 4The Hypervisor .......................................................................... 4Virtualization Makes Hardware More Important ................... 7The Virtualization Revolution in Hardware ............................ 7

Chapter 2: Putting Virtualization to Work . . . . . . . . . . . .11What Virtualization Delivers .................................................. 11Who Needs Virtualization Technology? ................................ 12AMD Opteron™ Processor: Moving toward Green ............... 15

Chapter 3: AMD Virtualization™ (AMD-V™) Technology . . . . . . . . . . . . . . . . . . . . . . . . . .19

Virtualizing Memory ................................................................ 20AMD-V™ Technology: Better Virtualization through

Better Allocation .................................................................. 21AMD Technology Helps Manage Memory ............................. 22Extending Virtualization to Devices ...................................... 24

Chapter 4: Virtualization for the Desktop and Client . . . . 29The Rise of Desktop Virtualization ........................................ 30High-Performance Graphics for Virtual Desktops ............... 30Client Virtualization: The New Kid on the Block.................. 31Hardware Extensions Aid Client Virtualization .................... 33Use Cases for Client Virtualization ........................................ 33

Chapter 5: Ten Great Reasons to Invest in Virtualization Hardware . . . . . . . . . . . . . . . . . . . . . .37

Create a Dynamic Data Center ............................................... 37Help Ease Power Consumption and “Stay Cool” .................. 38Provide Better Security ........................................................... 39Run Legacy Software on Non-Legacy Hardware .................. 39Develop and Test New Stuff Easily ........................................ 40Run Multiple Operating Systems on the Same Hardware ... 40Improve Scalability .................................................................. 41Enhance Your Hardware Utilization ...................................... 41Create a Manageable Upgrade Path....................................... 41Manage Outages (Expected and Unexpected) Dynamically ....42

02_644126-ftoc.indd iii02_644126-ftoc.indd iii 5/3/10 11:48 PM5/3/10 11:48 PM


Publisher’s AcknowledgmentsWe’re proud of this book; please send us your comments through our Dummies online registration form located at http://dummies.custhelp.com. For other comments, please contact our Customer Care Department within the U.S. at 877-762-2974, outside the U.S. at 317-572-3993, or fax 317-572-4002. For details on how to create a custom For Dummies book for your business or organization, contact [email protected]. For information about licensing the For Dummies brand for products or services, contact BrandedRights&[email protected].

Some of the people who helped bring this book to market include the following:

Acquisitions, Editorial, and

Media Development

Project Editor: Jennifer Bingham

Editorial Manager: Rev Mengle

Business Development Representative: Karen Hattan

Custom Publishing Project Specialist: Michael Sullivan

Composition Services

Project Coordinator: Kristie Rees

Layout and Graphics: Erin Zeltner

Proofreader: Jessica Kramer

Publishing and Editorial for Technology Dummies

Richard Swadley, Vice President and Executive Group Publisher

Andy Cummings, Vice President and Publisher

Mary Bednarek, Executive Director, Acquisitions

Mary C. Corder, Editorial Director

Publishing and Editorial for Consumer Dummies

Diane Graves Steele, Vice President and Publisher, Consumer Dummies

Composition Services

Debbie Stailey, Director of Composition Services

03_644126-flast.indd iv03_644126-flast.indd iv 5/3/10 11:48 PM5/3/10 11:48 PM


Introduction

Virtualization is a technical innovation designed to increase the level of system abstraction and enable IT users to

harness ever-increasing levels of computer performance.

At its simplest level, virtualization allows you, virtually and cost-effectively, to have two or more virtual comput-ing environments, running different operating systems and applications on one piece of hardware. For example, with virtualization, you can have both a Linux virtual machine and a Microsoft Windows virtual machine on one system. Alternatively, you could host a Microsoft Windows 95 desktop and a Microsoft Windows XP desktop on one workstation.

In slightly more technical terms, virtualization essentially decouples users, operating systems, and applications from the specific hardware characteristics of the systems they use to perform computational tasks. This technology promises to usher in an entirely new wave of hardware and software inno-vation. For example, and among other benefits, virtualization is designed to simplify system upgrades (and in some cases may eliminate the need for such upgrades), by allowing users to capture the state of a virtual machine (VM) and then transport that state in its entirety from an old to a new host system.

Virtualization is also designed to enable a generation of more energy-efficient computing. Processor, memory, and storage resources that today must be delivered in fixed amounts determined by real hardware system configurations will be delivered with finer granularity via dynamically tuned VMs.

Already, the combination of software-based virtualization and AMD’s hardware-based virtualization technology, AMD Virtualization™ (AMD-V™) technology, allows users to deploy computing resources in agile, efficient, and cost-effective ways. AMD is proud to play a pivotal role in virtualization and thereby help its customers extract ever-increasing value from their IT investments.

04_644126-intro.indd 104_644126-intro.indd 1 5/3/10 11:48 PM5/3/10 11:48 PM


Virtualization For Dummies, 2nd AMD Special Edition 2

About This BookVirtualization For Dummies, 2nd AMD Special Edition explains how virtualization works, how it can benefit your organization, and how the x86 architecture is engineered to better integrate computing hardware with virtualization software making virtu-alization faster and more reliable.

The contents of this custom book were provided by and pub-lished specifically for AMD.

Foolish AssumptionsAll authors make assumptions as they write their books. As we wrote this, a few things we assumed about you included the following:

✓ You’re an IT decision maker.

✓ You may be the executive officer of a company.

✓ You’re already familiar with computer “lingo.”

Icons Used in This BookIn the margins of this book, you find several helpful little icons that can make your journey a little easier:

This icon flags information that you should pay attention to.

This icon lets you know that the accompanying text explains some technical information in detail. You don’t need to know this stuff to get what you need from the book, but it may be interesting.

A Tip icon lets you know that some practical information that can really help you out is on the way. These tips can help save you time, effort, or money.

04_644126-intro.indd 204_644126-intro.indd 2 5/3/10 11:49 PM5/3/10 11:49 PM


Chapter 1

The Roots of VirtualizationIn This Chapter▶ Understanding virtualization

▶ Examining the virtualization ecosystem

▶ Looking at x86 server enhancements and multi-core processors

It seems like everywhere you go these days, someone is talking about virtualization. Technical magazines trumpet

the technology on their covers. Virtualization sessions are fea-tured prominently at technology conferences. And, predictably enough, technology vendors are describing how their product is the latest in virtualization.

Why VirtualizationTechnological evolution both drives, and is driven by, ever-increasing levels of abstraction in hardware and software architectures. High-level programming languages allow pro-grammers to implement software without having to pay too much attention to the operating systems on which it will run. One job of operating systems is to provide the abstractions that free programs from the complex and varied details needed to manage memory and input/output (I/O) devices.

The operating systems that provide this isolation must be totally cognizant of the hardware on which they reside. Details like MAC and IP addresses, SAN LUN assignments, physical memory configurations, processor counts, and system serial numbers become enmeshed within the OS

05_644126-ch01.indd 305_644126-ch01.indd 3 5/3/10 11:49 PM5/3/10 11:49 PM


Virtualization For Dummies, 2nd AMD Special Edition 4state at the time of system installation. This stateful informa-tion locks the OS to the specific hardware on which it was installed and complicates hardware fault recovery, system upgrades, and application consolidation.

Coming Up with the ConceptHardware and software architects realized that if they could abstract the hardware as seen by the operating system, then they could finesse the software’s view of the physical configuration on which it was installed. This approach is kind of like the black box model — the software doesn’t really need to know what’s going on inside the black box as long as computations sent into the box come back out with the correct results. The architects called their approach virtualization, and it turns out to be more complex than you might imagine.

Complexity arises because an operating system believes it owns all resources on the hardware on which it runs. More than that, an operating system doesn’t like to be fooled. This challenge is made more difficult still, given that the x86 architecture came into existence before notions of virtualization became common currency in mainstream computing.

The HypervisorServer virtualization enables you to consolidate many differ-ent types of workloads and operating systems onto virtual environments all running on a single hardware platform. Virtual servers or virtual machines are independent operat-ing environments that use virtual resources. One of the most common approaches to virtualization is to use hypervisor technology. Hypervisors use a thin layer of code in software to achieve fine-grained, dynamic resource sharing. Today’s hypervisors provide a high level of flexibility in how virtual resources are defined and managed and have become a favor-ite choice for server virtualization.



Chapter 1: The Roots of Virtualization 5Two types of hypervisors exist. The two types are discussed in the following sections.

Type 1 hypervisor solutionsYou will find that Type 1 hypervisors are typically the preferred approach for server consolidation because they can achieve higher virtualization efficiency than Type 2 hypervisors by dealing directly with the hardware. Type 1 hypervisors provide higher performance and efficiency, and use hardware assisted virtualization technology like AMD-V™ technology. Powering ultrathin notebooks to blade servers, all AMB processors shipped are designed to use AMD-V features.

Type 1 hypervisors use a thin layer of code to provide resource sharing within a single hardware platform. In other words, the hypervisor provides a standard emulated hardware environ-ment that the guest OS, sometimes referred to as the virtual machine (VM), resides on and interacts with. The VM encapsu-lates the guest operating systems and any applications running on them into a single entity that is isolated from the underlying hardware. It is because of this encapsulation that the VM can be migrated from one physical machine to another without any service interruption.

Not only does this approach support running multiple VMs on the same hardware, it can also support multiple VMs running different types and/or versions of operating systems (for example, completely different operating systems like Windows and Linux can be run simultaneously on the same physical server).

Type 1 hypervisor solutions are often used for server con-solidation to achieve higher levels of resource utilization. Software development and quality assurance environments can also benefit greatly from this type of virtualization due to the ability to allow a number of different operating systems to be run simultaneously. This can facilitate parallel develop-ment or testing of software in a number of different operating system environments, thus leading to quicker, and perhaps more efficient, testing phases.



Virtualization For Dummies, 2nd AMD Special Edition 6Early Type 1 hypervisor solutions experienced some perfor-mance degradation due to the overhead of the virtualization software, however, with the introduction of hardware-assisted virtualization that is built into some of today’s multi-core processors, and platform improvements, such as expanded memory and I/O capabilities, this performance degradation has been minimized. In fact, there is evidence that some native applications may run as well, if not better, in a virtual environment due to these improvements.

Operating system virtualization — Type 2 hypervisorType 2 hypervisors run on a host operating system that pro-vides virtualization services such as I/O device support and memory management. These services give an application the illusion that it is (or they are, if there are multiple applica-tions) running on a machine dedicated to its use. The key thing to understand is that, from the application’s execution perspective, it sees and interacts only with those applications running within its virtual OS, and interacts with its virtual OS as though it has sole control of the resources of the virtual OS. However, it can’t see the applications or the OS resources located in another virtual OS.

Virtualization solutions that use a Type 2 hypervisor are also referred to as operating system (OS) virtualization, and in some environments are called containers.

This approach to virtualization is extremely useful if you want to offer a similar set of operating system functionalities to a number of different user populations while using only a single machine. This is an ideal approach for Web hosting compa-nies: They use container virtualization to allow a hosted Web site to “believe” it has complete control of a machine, while in fact each hosted Web site shares the machine with many other Web sites.

There are some limitations to operating system virtualization, though. First and foremost, this approach typically limits oper-ating system choice. Containerization usually means that the containers must offer the same operating system as the host



Chapter 1: The Roots of Virtualization 7OS and even be consistent in terms of version number and patch level. As you may imagine, this can cause problems if you want to run different applications in the containers, because applications are often certified for only a certain OS version and patch level. Consequently, operating system virtualization is best suited for homogenous configurations.

Virtualization Makes Hardware More Important

Even though virtualization is a software technology, it has the effect of making hardware more important. Removing lots of servers and migrating their operating systems to virtual machines makes the remaining servers that support all those virtual machines vital to running your business.

For example, in an organization that has a “one application, one server” environment, if a single server goes down, this inconveniences a single user population. However, virtualiza-tion is very different. If each server supports multiple virtual machines, then a server that goes down would inconvenience multiple user populations. Thus, the hardware can take on greater importance in a virtual environment.

The greater importance placed upon hardware by virtualiza-tion technologies is one reason why hardware manufacturers like AMD have introduced virtualization-ready technologies like AMD-V technology. AMD-V technology can be found in all AMD multi-core processors, and other hardware manufacturers have introduced similar enhancements to support virtualization implementations.

The Virtualization Revolution in Hardware

Because virtualization inevitably leads to more apps running on a single hardware instance, hardware becomes more impor-tant. And hardware has responded. Well, actually, hardware manufacturers have responded, but you get the point.



Virtualization For Dummies, 2nd AMD Special Edition 8The virtualization-enabling hardware developments break into two categories:

✓ Virtualization extensions to processors and associated low-level hardware like network interface cards (NICs). These extensions allow virtualization to integrate more closely with the underlying hardware in order to improve performance and security. For more on these hardware changes, see Chapter 3.

✓ Form factor changes to servers to support virtualiza-tion workloads more effectively. Originally, standard 1U servers (sometimes called pizza boxes) were used as virtualization platforms. These servers typically had one processor, up to 16GB of memory, and a limited number of NICs — a design that makes sense in a one application, one server world. In the new virtualization-oriented world, though, every one of these capabilities turned into a bot-tleneck: To run large virtualization workloads, these serv-ers had too little processing power, not enough memory, and nowhere near enough network connections. So server manufacturers have transformed their server designs to better support virtualization with these design extensions:

• Multi-core processors. Today’s multi-core proces-sors help to improve utilization and increase VM density by allowing many virtual machines to share a single physical server. Additionally, today’s multi-core processors can be more energy efficient than single core processors, depending on their usage and configuration, potentially providing further energy cost savings.

• Multi-socket servers. Today, single-socket servers are passé for virtualization. Servers with two or four multi-core processors, sometimes referred to as 2P or 4P systems, are the norm.

• Big memory servers. Of all the bottlenecks in virtu-alization, memory is typically the first encountered. Solution? More memory. Manufacturers are now releasing servers with memory capacities of up to half a terabyte of memory, and, as with all things hardware, you can expect to see memory capacity climbing in the future.



Chapter 1: The Roots of Virtualization 9 • Blade servers. In yet another scheme to pack more

processing into less room, server manufacturers have pushed all this hardware goodness into blade servers — servers that contain blades (actually, an entire server contained on a small plug-in board) are slotted into a multi-U chassis. Instead of 4 units in a rack holding four computers, a blade server can contain 16 or even 32 blade computers in the same space. Bottom line, more server for a given amount of rack space, and more virtualization for a given amount of rack space.






Chapter 2

Putting Virtualization to Work

In This Chapter▶ Understanding what capabilities virtualization offers

▶ Looking at the benefits of implementing virtualization

▶ Going green

This chapter explains what capabilities virtualization offers and how it can improve day-to-day operations in data

centers. Virtualization is far more than just the latest shiny thing from Silicon Valley — its capabilities have revolutionized data center operations and provided enormous cost savings across companies and industries. Read on to learn more about the benefits of virtualization.

What Virtualization DeliversAmong other things, virtualization allows you to:

✓ Consolidate workloads to help reduce hardware, power, and space requirements.

✓ Run multiple operating systems simultaneously, as an enterprise upgrade path, as a way to leverage the advan-tages of specific operating systems, or for whatever reason you can imagine.

✓ Run legacy software on hardware that is newer, more reliable, and likely more power-efficient.



Virtualization For Dummies, 2nd AMD Special Edition 12 ✓ Dynamically migrate workloads to provide fault tolerance.

✓ Provide redundancy to support disaster recovery.

In the rest of this chapter, we help you understand these and other benefits of creating and employing virtualized environments, as well as how many of the risks that used to be associated with virtualization are being eliminated by hardware that better supports virtualization software.

Who Needs Virtualization Technology?

Today, some data centers may have machines running at only 10 or 15 percent of total processing capacity. In other words, 85 or 90 percent of the machine’s potential is unused. However, a lightly loaded machine still takes up room and draws electricity, so the operating cost of today’s under-utilized machine can be high.

It doesn’t take a rocket scientist to recognize that this situ-ation is a waste of computing resources. And, guess what? With the steady improvement in performance characteristics of computer hardware, this year’s machines are better than last year’s, next year’s machine will be better than this year’s, and so on, for the foreseeable future. Obviously, there ought to be a more efficient means of matching computing capacity with load. Fortunately, virtualization can provide that means. Virtualization enables a single piece of hardware to seam-lessly support multiple systems, which means organizations can raise their hardware utilization rates dramatically.

Enabling the dynamic data centerThe business world has undergone an enormous transforma-tion over the past 20 years. Business process after business process has been captured in software and automated, moving from paper to electrons.



Chapter 2: Putting Virtualization to Work 13The rise of the Internet has exponentially increased this trans-formation. Using the worldwide connectivity of the Internet, companies want to communicate with customers and part-ners in real-time. Naturally, this too has accelerated the move to computerized business processes.

Over time, organizations have adopted a variety of hardware and software systems throughout the enterprise. Today’s complex data-server environments have grown exponentially, leaving companies with larger and more rigid data centers. Not only can it become cost-prohibitive to administer these data centers, it can also be difficult to achieve optimal value, return on investment (ROI), and return on assets (ROA).

Virtualization enables organizations to make more effective use of their computing resources and to move legacy software to newer, more efficient hardware platforms. The ability to run different operating systems and applications on the same physical server lets organizations consolidate the workload placed on servers. Virtualization partitions a server into several virtual machines, each able to run its own separate operating system and application environment. This moves businesses away from a “one server, one application” model toward an infrastructure that lets a business manage its servers across a heterogeneous environment far more effectively.

System administration costs mountComputers don’t operate all on their own. Every server requires TLC from system administrators. Common system administration tasks include monitoring hardware status, replacing defective hardware components, installing operat-ing system (OS) and application software, installing OS and application patches, monitoring critical server resources like memory and disk use, and backing up server data to other storage mediums for security and redundancy purposes.



Virtualization For Dummies, 2nd AMD Special Edition 14As you can imagine, these tasks can be pretty labor intensive. System administrators — the people who keep the machines humming — don’t come cheap. And, unlike programmers, system administrators are usually co-located with the servers, because they need to access the physical hardware.

Virtualization enables an organization to streamline and automate a host of complex processes, respond to changing business requirements, and achieve the flexibility required to compete in today’s global economy. This can often result in a reduction in the costs associated with the management and administration of the data center because for many organiza-tions, system and network administration is an expensive and unwieldy task. Consider that, as the computing environment becomes more complex, IT staffing needs and tasks will likely increase in both number and complexity. However, virtualiza-tion can tip the balance back in favor of the enterprise. A well-planned virtualization strategy can reduce the time it takes for IT staff to manage the server infrastructure. This leaves hard-ware specialists, programmers, and other professional staff free to handle more strategic work.

Green initiatives seek better energy efficiencyYears ago, power costs factored into strategic thinking at about the same level as what brand of soda to keep in the vending machines. Companies generally assumed that electri-cal power would be cheap and endlessly available, so it was not a major factor in strategic plans.

The assumption regarding availability of reliable power was challenged during the California power scares of a few years ago. Although later evidence caused re-evaluation of whether there was a true power shortage, the events caused many companies to consider whether they should look for ways to be less power dependent.

Furthermore, the impact of the green revolution has encour-aged many companies to look for ways to reduce the amount of energy they consume.



Chapter 2: Putting Virtualization to Work 15One way in which companies have done this is by turning their focus to their data centers. To show the level of concern about the amount of energy being consumed in data centers, consider these facts:

✓ According to a study performed by researchers from the Lawrence Berkeley National Laboratory and facts pub-lished by the U.S. Department of Energy, data centers in the U.S. doubled their energy consumption between 2000 and 2006, and are likely to double that again by 2011. In addition, the U.S. Environmental Protection Agency (EPA) found that in 2006, cooling costs for data centers accounted for 1.5 percent of the total energy consumed in the U.S.

✓ Based, in part, on the results of an earlier study by this group, the U.S. EPA has convened a working group to establish standards for server energy consumption and plans to establish a new “Energy Star” rating for energy-efficient servers.

The cost of running a large number of servers, coupled with the fact that many of the machines filling up data centers are running at low utilization rates, means that a company that implements virtualization technologies may be able to reduce the total number of physical servers, and thereby reduce its overall energy costs. This is another way in which virtualiza-tion can help.

AMD Opteron™ Processor: Moving toward Green

A leading factor in why companies are moving to virtualiza-tion is to help reduce energy consumption. Big data centers can use incredible amounts of power, and anything that helps reduce power is welcome.

Although it may seem that the power savings for one chip would be relatively insignificant, keep in mind that today’s data centers can contain thousands of machines. Even a virtu-alized data center, where many physical machines have been converted to guest virtual machines, can contain hundreds of physical servers.



Virtualization For Dummies, 2nd AMD Special Edition 16AMD is doing its part to help reduce power consumption, but not at the cost of decreased performance. AMD has included a number of features within its AMD Opteron processors, including AMD-V™ technology, hardware assisted virtualiza-tion technology, and AMD-P, a suite of power management capabilities that can dynamically minimize power usage of the overall processor, individual cores, and the logic within each core. For more info on AMD-P, see Table 2-1.

Table 2-1 The AMD-P Suite of Power Management Capabilities

Capability Benefit

AMD CoolCore™ Technology

Can reduce energy consumption by turning off unused parts of the processor.

AMD Smart Fetch Technology

Helps reduce power consumption by allowing idle cores to enter a halt state, causing them to draw even less power during processing idle times, without compromising system performance.

Independent Dynamic Core Technology

Enables variable clock frequency for each core, depending on the specific performance requirement of the appli-cations it is supporting, helping to reduce power consumption.

Dual Dynamic Power Management (DDPM) Technology

Provides an independent power supply to the cores and to the memory controller, allowing the cores and memory controller to operate on differ-ent voltages, depending on usage.

AMD PowerCap manager Provides the ability to put a cap on the P-state level of a core via the BIOS, helping to deliver consistent, predict-able power consumption of a system.

AMD C1E A sleep state invoked when processor cores are at idle that can equate to significant power savings in the data center.



Chapter 2: Putting Virtualization to Work 17

AMD-P enables the processor to provide the needed perfor-mance while helping reduce power consumption both directly and indirectly: direct power consumption by the processors, and indirect power consumption required to deal with the heat thrown off by those processors. Together, those tech-nologies can help enable cutting-edge energy efficiency.

By running AMD Opteron processors with AMD-V technol-ogy and AMD-P in a virtualized environment, users can save energy in two ways:

✓ By dynamically adjusting to processing demands, each machine can tailor the amount of energy needed to the specific job. Thus, users can save on overall energy con-sumption by using energy on an as-needed basis.

✓ Using less energy means less heat in the chip (chips generate heat as they process information). By reduc-ing the heat generated by the machine, users will need less air conditioning in the data center, further reducing energy consumption (and saving even more money on energy costs!).

So not only can AMD Opteron processors help to make your virtualization systems run more effectively, they can make your data center run more efficiently.






Chapter 3

AMD Virtualization™ (AMD-V™) Technology

In This Chapter▶ Virtualizing memory

▶ Digging into AMD Virtualization technology

▶ Examining I/O virtualization technology

Virtualization originally was a software innovation that required a very smart hypervisor to mediate resource

contention between physical hardware resources like the computer’s processor, memory, network, and storage. Even though the virtualization software was very smart, naturally people began to think how much better it could be if it were only . . . higher performance! And so AMD set to work to see how to make virtualization even more efficient. Their innova-tions in areas like AMD Virtualization (AMD-V) technology and I/O virtualization with IOMMU, offer better performance due to hardware optimization.

As individual virtual machines improve, more of them can be supported on a given piece of hardware; in other words, if virtual machine performance goes up, more virtual machines can be squeezed onto one server, thereby achieving higher virtual machine density.

Virtual machine density refers to the ratio of virtual machines to physical machines. The higher the number of virtual machines that can be supported on a physical system, the higher the virtual machine density. Put another way, the higher the den-sity, the lower the number of physical machines required to run an organization’s virtual systems.



Virtualization For Dummies, 2nd AMD Special Edition 20Performance can be another key element when it comes to virtual machine density, but raw performance doesn’t always result in optimal throughput and energy efficiency. For vir-tualization solutions running many disparate workloads on a single physical server, it is often important to balance raw per-formance and energy efficiency — also known as performance-per-watt — as well as factoring in overall system price, when assessing the bottom line of any virtualized solution.

Virtualizing MemoryLong before computer scientists came up with the notion of virtualizing an entire system, architects had already invented techniques to virtualize memory management. The Atlas com-puter at the University of Manchester was the first system to incorporate virtual memory technology. Virtual memory technology lets a system with a limited amount of physical memory look much larger to application software. To create this illusion, the OS stores the full memory image of the appli-cation and its data on the system’s hard drive, and transfers required pieces of this image into the system’s DRAM memory as the program executes.

To translate the virtual addresses seen by each application into physical DRAM memory addresses, the system relies on a map (known as a page table) that contains references linking chunks of virtual memory to real memory. Contemporary x86 processors include hardware features known as translation look-aside buffers (TLBs) that cache the translation refer-ences for recently accessed chunks of memory, thus speeding up the process. TLBs play a role in almost all memory refer-ences, so the manner in which they perform their translations can play a significant role in determining overall system performance.

Architects soon learned that TLB design can seriously impact multitasking systems operations. Most tasks in such systems have unique page tables, which forces the operating system to reset (or, more colorfully, “flush”) the TLB each time it switches from one task to another. Then, as the new task executes, its page table entries fill up the TLB, at least until the next task switch. This constant flushing and reloading can really eat into performance, especially if each task runs for only a few milliseconds before the next switch.



Chapter 3: AMD Virtualization™ (AMD-V™) Technology 21

AMD-V™ Technology: Better Virtualization through Better Allocation

Clearly, a more efficient approach to managing hardware resources would help overall virtualization performance. To that end, AMD introduced AMD-V technology with its Dual-Core AMD Opteron™ processors in 2006. AMD-V technol-ogy is designed to improve virtualization performance with processor-based hardware enhancements.

AMD-V technology also leverages Direct Connect Architecture with an integrated memory controller to provide fast and effi-cient memory management.

So how did this play out with respect to that pesky memory management? AMD processor architects added a new, virtual machine (VM) specific tag called an address space identifier (ASID) to the TLBs in the AMD Opteron processors. This concept is known as a tagged translation look-aside buffer, or tagged TLB.

Each VM has a unique ASID value, known only to the hyper-visor and the TLB hardware. The ASID is invisible to the guest OS, thus eliminating the need to modify the guest, preserving the virtual illusion and avoiding any performance degrada-tion. Figure 3-1 illustrates the tagged TLB concept.

VM 1 runs on the CPU andloads additional data frommemory

As VM 3 takes control andloads its data, other TLBdata remains

So when VM 1 takes controlback the data it needs isthere … resulting in better performance

VM 1 VM 2

AMD Non-AMD

VM 3

VM 1

VM 3

VM 1

VM 4 VM 5 VM 6

Hypervisor (VMM)Hypervisor (VMM)

HT 3

HT 1

HT 2

AMD Opteron™Tagged TLBCache lines

Mem

oryController

Legacy x86ArchitectureUn-Tagged TLB

Cache linesFront-side Bus

Figure 3-1: Tagged translation look-aside buffer.




AMD Technology Helps Manage Memory

One of the most important tasks for a hypervisor is memory management. The ability to keep track of the memory for indi-vidual processes within a virtual machine is critical for virtu-alization. Even more important is to ensure that each virtual machine’s overall memory is managed. The latter task may be referred to as keeping track of virtual machine state — the settings of all critical system variables at each moment in time.

Shadow pagingSoftware-only virtualization solutions use a technique called shadow paging. Shadow paging uses stored information about the physical location of guest memory in shadow page tables and manages these structures in the hypervisor. Under shadow paging, the hypervisor intercepts guest page table updates to keep the shadow page tables in synch with the guest page tables.

Starting with the first Quad-Core AMD Opteron processors, AMD implemented a hardware optimization to memory manage-ment called rapid virtualization indexing (RVI). RVI translates memory addresses from virtual to physical, but unlike shadow page tables, which perform these translations in software, RVI performs the memory translation in the CPU (see Figure 3-2).

Understanding the flow of memoryTo understand what rapid virtualization indexing accom-plishes, it’s important to understand the flow of memory in a virtualized environment:

1. The virtual machine operating system has its own vir-tual memory that enables the system to “pretend” its total available memory is larger than is really available. Page tables swap memory back and forth onto disk to enable this. (This is called guest virtual memory.)



Chapter 3: AMD Virtualization™ (AMD-V™) Technology 23 2. The virtual machine has actual memory that this vir-

tual memory is swapped into and out of as needed. In a virtualized environment, this actual memory is managed by the hypervisor and is, in fact, also virtual. (This is called guest physical memory.)

3. The hypervisor itself manages a pool of memory that may be larger than the physical memory available on the underlying server. It has pages that it swaps back and forth with the disk to support this virtual memory.

4. Finally (at last!) there is the physical memory on the hardware system, which is where actual processing occurs. (This is called host physical memory.)

Rapid virtualization indexing (RVI) provides this hardware mechanism to determine the physical location of guest memory by walking an extra level of page tables — called nested page tables.

Guest Linear

Hypervisor translates a page in guest virtual address space to machine physical space through a two-level translation

– First, map guest virtual address to guest physical address– Then, map guest physical address to machine physical address

0

Guest Physical

gCR3

page

d by

paged by

page

d by

page

d by

nCR3

nCR3

TLB

Entry

the VMM’s CR3

gCR3

CR3 (used by VMM)

0

VMM Host Linear

0

System Physical

0

PTnPT gPT

gPT

nCR3

Figure 3-2: Design of AMD’s Rapid Virtualization Indexing.



Virtualization For Dummies, 2nd AMD Special Edition 24In a native system, the operating system maintains a mapping of logical page numbers to physical page numbers in page table structures. When a logical address is accessed, the hard-ware walks these page tables to determine the corresponding physical address. For faster memory access, the x86 hardware caches the most recent mappings in its translation lookaside buffer (TLB).

In a virtualized system, the guest operating system maintains page tables just like in a native system, but the hypervisor maintains an additional mapping of the guest physical page numbers to host physical page numbers in the shadow page table structures.

RVI enables the hypervisor to maintain this guest-physical to host-physical mapping in a second level of page tables, called nested page tables in hardware.

When a logical address is accessed, the hardware walks the guest page tables as in the case of native execution, but for every guest physical page accessed during the guest page table walk, the hardware also walks the nested page tables to determine the corresponding host physical page address. This eliminates the need to maintain shadow page tables and synchronize them with the guest page tables.

Overall, this can help reduce the overhead of accessing memory, thereby increasing system performance. The extent of this performance increase is somewhat dependant on the type of workload the virtual machine is executing; memory-intensive applications see more performance improvement than applications that aren’t heavily dependent on memory access. But by and large, moving the memory access func-tions into hardware can improve performance, which in turn can improve virtual machine density.

Extending Virtualization to Devices

With AMD-V technology, AMD has created ways to improve the performance of network, storage, and graphics in a virtu-alized environment. AMD has also introduced technology for



Chapter 3: AMD Virtualization™ (AMD-V™) Technology 25I/O virtualization which resides in the AMD chipset. AMD-V uses an I/O Memory Management Unit, often called IOMMU, to control how a device accesses memory. The IOMMU also provides a mechanism to isolate memory accesses by these devices, thereby helping to improve the security in a virtual-ized environment

Locating memoryIn order for information to flow back and forth to these input/output devices (this is where the IO in IOMMU comes from), it first must be moved to memory that is accessible by the I/O device. Each I/O device connected to a computer has its own specific location in the system memory. The operating system knows just where that memory is, so when it wants to send data, say, across the network, it transfers data from the proces-sor’s memory to the I/O device’s memory, where the I/O device can access it and send it on its merry way. When data returns, the I/O device puts it into its assigned memory location, where the OS grabs it and transfers it into the processor’s memory, where it can be processed.

This gets complicated in a virtualized environment. The guest operating system directs I/O-bound data to where it thinks the I/O device can grab it, but the virtualization hypervisor cleverly intercepts the attempt by the guest OS to write to physical memory and maps it through its own memory and then onto the actual physical memory that the I/O device is attached to.

The hypervisor must keep track of and map all the different guests’ virtual I/O memory locations and constantly swap the virtual representations of the guests’ I/O memory locations into the actual physical I/O memory. As should be pretty clear, this requires really, really smart hypervisor software so that all I/O interactions can be kept straight. After all, you wouldn’t want your CRM system to be reading data from your DNS (domain name service) system, would you?

Of course, one wouldn’t use the term “kept straight” in a complex computer science topic like virtualization — it sounds so . . . casual. By leveraging AMD’s IOMMU, systems ensure data integrity (“keep data straight”) as well as enforce security, since the functionality isolates individual VM I/O from one another. One VM can’t access the I/O memory locations of another VM.



Virtualization For Dummies, 2nd AMD Special Edition 26With I/O memory mapping, the hypervisor must perform this task efficiently — very efficiently. Doing so is particularly important because I/O is critical for overall system performance — after all, no computer operates without accessing data on a hard drive. In today’s computing world, applications typically interact with users or other systems across a network. So I/O performance is an important area that really needs optimization to ensure acceptable perfor-mance, and, of course, good virtual machine density.

Chipsets that work with youThe processor and the chipset really are partners and must be designed to work together. AMD has delivered the AMD SR56x0 chipset, which integrates chipset-resident functions into its AMD Opteron processor-based platforms and is enabled with I/O virtualization technology (IOMMU) for robust support for memory- and I/O-intensive environments.

Understanding IOMMUAMD has taken the lead in moving functions originally per-formed by the hypervisor in software into hardware, and IOMMU is no different. In fact, IOMMU takes advantage of some of the same architectural approaches used in shadow page tables.

In essence, IOMMU subdivides the I/O memory associated with an I/O device and allows a hypervisor to create dedi-cated subsections of the memory that may be assigned to virtual machines (see Figure 3-3). In this way, each virtual machine has a section of memory dedicated to its I/O use, which means the hypervisor can set up the original dedicated connection, and then let the virtual machine communicate directly with the I/O device without needing to be involved. This can reduce the software processing overhead and improve performance.

Subdividing the memory assigned to an I/O device presents some challenges: how to keep track of each of the subdivisions, how to ensure that the I/O device places the appropriate data into the right memory subdivision, and how to ensure that only the right virtual machine accesses that subdivision to get its assigned data.



Chapter 3: AMD Virtualization™ (AMD-V™) Technology 27

CPU

HyperTransport™technology link

HyperTransport™technology link

DRAM

PCIe™ bus

Memory Controller

IOMMU

I/O Hub

TLB IOMMU

I/O Hub

TLB

LPC

PCI

SATAIntegratedPeripherals

I/ODevice

I/ODevice

IOTLB

I/O Device

PCIe™ busPCIe™ bus

Figure 3-3: How IOMMU works.

How does IOMMU meet this challenge? Remember our old friend the translation look-aside buffer (TLB)? Well, IOMMU takes advantage of TLBs to isolate the various subdivisions of the I/O memory. Furthermore, it uses a Domain ID to keep track of the assignments between particular memory areas and the guest virtual machines on the system.

By implementing IOMMU, a hypervisor can reduce the number of steps required to get data from a virtual machine out to a physical I/O device. By bypassing some steps, the system will have to execute fewer software instructions, which can raise system performance, and, ultimately, drive higher virtual machine density.






Chapter 4

Virtualization for the Desktop and Client

In This Chapter▶ Examining how virtualization is key to the future of computing

▶ Tracing the rise of client device virtualization

▶ Looking at different types of client virtualization

▶ Understanding how hardware helps client virtualization

It may seem like all the virtualization fun is centered on servers located in the data center. But have no fear!

Virtualization has come to client devices (more informally known as desktops and laptops), and can offer the same types of benefits that server virtualization offers — and a few others besides.

The evolving concept of a virtualized desktop is evidenced by the recently formed Virtual Desktop Infrastructure Alliance that helps IT administrators create and manage “desktop vir-tual machines” on servers within data centers. End-users may access these desktop environments at any time and from any place, using thin client devices (or thin client access utilities on more fully configured systems). Even old, underpowered systems can use the Remote Desktop Protocol (RDP) to access more powerful virtual PC desktops loaded with up-to-date software versions.

This approach to client deployment can lead to lower sup-port expenses as well as hardware acquisition costs because the virtual PCs reside on centralized servers in a managed IT environment. This can eliminate the need to visit the client’s actual desktop system for most maintenance activities, which can help reduce IT personnel costs and overhead.




The Rise of Desktop Virtualization

Client virtualization is a growing trend — but it’s based in technology that has been available for a number of years. The ability to display a desktop remotely, known as desktop virtualization, has been part of Windows for at least a decade, but most organizations found the capability too limited for general use. Two primary problems held back the spread of desktop virtualization:

✓ Poor overall performance. Trying to send an entire desktop across a network is a challenge, and the early protocols to support desktop virtualization weren’t efficient enough to provide sufficient performance, resulting in slow desktop refreshes, jerky motions — in short, unsatisfactory experiences.

✓ Poor graphics performance. Even after more efficient protocols became available, graphics-rich programs still suffered because of the difficulty of transmitting rapidly changing output.

However, the rise of server virtualization, which made hosting desktops in the data center easy and efficient, enabled new developments in desktop virtualization.

High-Performance Graphics for Virtual Desktops

So what about graphics performance in this new world of Virtual Desktops? Users demand advanced graphics features like standard USB connectivity, multi-monitor support, and complex streaming and 3D support. Well, here is another example of where AMD is leading the charge when it comes to graphics and virtualization.

A relatively new protocol that enables remote access to high-performance graphics applications in the data center is evolv-ing called PC-over-IP (PCoIP). The PCoIP protocol compresses,



Chapter 4: Virtualization for the Desktop and Client 31encrypts, and encodes the entire computing experience at the data center and transmits only the changing display pixels across a standard IP network to stateless PCoIP technology enabled devices. This greatly reduces the amount of network bandwidth required to feed the end-user’s thin client system while still providing a constant and robust graphics update. PCoIP technology is delivered in both hardware and software based implementations, and both types of solutions are now supported by AMD’s FirePro™ line of ATI Graphics cards.

An example of these devices is the ATI FirePro RG220 card. This card supports the PcoIP protocol and can be installed in servers residing in the data center. Dedicating a graphics card per client, along with support for PcoIP, can provide high-quality 2D, 3D, and complex graphics support.

Some AMD FirePro graphics cards, working in conjunction with AMD Virtualization™ (AMD-V™) technology, can support a new feature called direct mapping. Direct mapping is a way in which a virtual machine (VM) can be directly mapped to a graphics driver. This now gives each VM its own fully dedi-cated graphics card to use for full, no-compromise graphics performance. Direct mapping coupled with multiple FirePro RG220s allows for a single server to create multiple virtual machines with full graphics performance and full multiple remoting connections.

Client Virtualization: The New Kid on the Block

One drawback to desktop virtualization is obvious: To use it you must be connected to the network. What about virtualiza-tion in environments in which you aren’t connected to the network? A new form of client-oriented virtualization, termed client virtualization (very creative, eh?) has sprung into being and is being widely deployed.

In this form of virtualization, rather than the client virtual machine residing in the data center, it sits on the client device and can be used locally — that is, without a network connec-tion to the data center.




Understanding client virtualizationJust as server virtualization abstracts an operating environ-ment (a combination of an operating system, software, and data into a virtual machine), client virtualization does the same thing — just for an end-user device like a desktop or laptop computer.

Client virtualization abstracts dependence on physical resources so that a complete operating environment may be encapsulated in a virtual machine — thereby enabling one client device to support multiple virtual machines, which makes client virtualization a very interesting proposition, indeed.

What client virtualization providesClient virtualization allows a single physical device to support multiple operating environments. Each operating environment is comprised of an operating system, application software, and data associated with that virtual machine.

With client virtualization, one can support several different virtual machines on that one device.

Examples of the types of virtual machines that can be hosted on one device include:

✓ Identical virtual machines. Suppose you want to offer users the ability to run different versions of an applica-tion software package. You could provide several differ-ent client machines, but that turns a lightweight laptop into a burden of several different machines — a real drag. Instead, you put several different virtual machines onto one laptop and, voila, system flexibility without back strain!

✓ Virtual machines with different operating system ver-sions. Suppose you want to run the same application soft-ware on different versions of the same operating system



Chapter 4: Virtualization for the Desktop and Client 33to test compatibility. See the previous paragraph — client virtualization allows system flexibility and your back still stays unstrained!

✓ Different operating systems on the same physical device. There are many scenarios where you might want to run, say, Windows and Linux on the same device. Before virtualization, there was no way to accomplish this. Even today, with Type 2 virtualization, laptops strain to support virtualization.

Hardware Extensions Aid Client Virtualization

Given that most laptops don’t offer the same kind of form factor developments outlined in Chapter 1 of this book, highly efficient virtualization is a must for successful client virtual-ization. Because most laptops are limited to 4 GB of memory, hardware extensions that help to improve efficiency and performance are a real help. In fact, many client virtualization solutions require hardware-assisted technology.

AMD long ago recognized virtualization on the client and the desktop and has included AMD-V technology in all the latest AMD processors for server, desktop, and mobile platforms.

Use Cases for Client Virtualization

The flexibility client virtualization offers simplifies many use scenarios. The following sections discuss some common use cases for client virtualization.

There are undoubtedly more client virtualization benefits that will be discovered in the future as the technology is more widely adopted. It’s safe to say, though, that the established practices associated with personal computing are about to undergo a transformation, leading to more efficiency and lower costs.




Software engineeringSoftware engineers traditionally face a difficult work situation. Many software applications require multiple systems, one running Windows, another Linux, yet another operating as a backend Web server. Each of these systems may require two or more versions, reflecting different versions of operat-ing system, application software, and configurations. Trying to keep these installed and configured properly in a tradi-tional environment hampers productivity, because lots of time is spent in installing and configuring different software products — not to mention tying up lots of expensive equipment.

With client virtualization, a software engineer can have many different virtual machines stored on a client machine’s disk, and bring up only the necessary ones to program or test a specific application topology. Instead of trying to keep track of many different pieces of hardware, the necessary virtual machines can be brought up in minutes. There’s another advantage, too. In today’s world of telecommuting and travel, software engineering’s traditional practice of using hardware for all the different parts of an application is unworkable; with client virtualization, a software engineer can be productive anywhere. Magic!

Software testingJust as software engineers can be relieved of a great deal of repetitive work when virtual machines replace physical machine, so too can software testers. In fact, software testers probably realize even more benefits because of the highly dynamic application configurations typical of software testing.

Another benefit of client virtualization for software testing is the support it offers for reproducing problems. In the past, a software tester who observed a bug would need to describe how the system was configured and what steps were taken with the software. With client virtualization, the virtual machine in which the problem was observed can be stored and provided to the software engineering team, which can immediately observe the problem without trying to reproduce it.



Chapter 4: Virtualization for the Desktop and Client 35

Mixing personal and work environmentsIt’s no secret: Folks live online today. Each of us uses social media applications, stores digital media like photos and music, runs personal apps — they’re all part of our modern digital life.

Unfortunately, online activity can cause problems. If you’re using your business-issued laptop, loading non-approved software can be against company policy. Even worse, your personal software may have operational conflicts with software necessary to your business activities. This, of course, doesn’t even address the question of whether you want your personal work intermixed with your business activities.

Most companies want to have a standard “image” of an approved combination of operating system, applications, and configurations. Allowing personal applications to be installed or Web sites accessed can compromise business application stability and expose the company to risk.

Client virtualization, with the ability to support multiple virtual machines can mitigate these problems. By “locking down” the official business virtual machine and offering a second virtual machine for personal use, companies can avoid issues of application conflict or corporate liability. Likewise, individual users can use a personal virtual machine during non-business hours, secure in the knowledge that personally useful applications can be operated with no danger of conflict with the business virtual machine.

TrainingTraining departments have a common problem: Every student requires a fresh environment to interact with, and setting up a fresh environment for every student is expensive and time-consuming.



Virtualization For Dummies, 2nd AMD Special Edition 36Client virtualization can reduce the amount of time spent in this repetitive process, because a fresh virtual machine with a software configuration set up for the student to begin work can be created and deployed at will. One virtual machine, configured properly, can be installed on every student’s PC, saving enormous amounts of work and, just as important, avoiding the inevitable mistakes that can occur when a manual process is carried out time and again.

System migrationLast — but not least — is the way that client virtualization can assist in system migration. Most corporate environments replace PCs every three or four years. Each of these replace-ments requires a costly backup and data migration of the existing client machine (not to mention these migrations often lose personally installed applications or data — see the “Mixing personal and work environments” section).

Instead of this costly process, imagine how easy a system migration can be if all that is required is for a virtual machine on the old system to be backed up to a central server and then deployed onto the new client machine.



Chapter 5

Ten Great Reasons to Invest in Virtualization Hardware

In This Chapter▶ Getting more secure

▶ Saving money

▶ Improving scalability

It’s here — the end of the book. And like all For Dummies books, this one includes a chapter listing ten things worth

listing.

In this chapter, we’ve collected ten great reasons to use hardware virtualization in your technology space.

Create a Dynamic Data CenterVirtualization helps you break the “one application per server” model, while defining your data center by logical boundaries rather than physical boundaries.

To date, most data centers operate with excess capacity so they can handle increases in demand as and when they arrive. However, when systems sit idle, they gobble up power, produce heat, take up space, and waste money.

In a virtualized data center, as demand for a particular appli-cation increases, you may be able to move those virtual machines (VMs) to a physical server with excess capacity to handle the workload. Conversely, when demand decreases, you simply reallocate your virtual resources accordingly.



Virtualization For Dummies, 2nd AMD Special Edition 38And because workloads increase and decrease all day long at undetermined times, each of the various VMs on a physical server can take advantage of any extra capacity without direct human intervention. The result is that you can build a data center with fewer total resources, because resources can be used as needed instead of being dedicated to single applica-tions or sitting and waiting until the next spike in demand.

At its core, virtualization isn’t really about virtual anything — it’s about the dynamic allocation of resources. Yes, the basic unit of virtualization is a virtual machine, but it’s what a vir-tual machine does that is so useful.

Virtualization enables you to allocate resources where you need them, when you need them, from an aggregate pool of resources. You can look at the aggregate demand of your envi-ronment, build out sufficient capacity to handle that demand, and then pull as necessary from that capacity and be able to handle multiple operating systems, servers, and applica-tions. This can save money on your total hardware investment (including potential power, heat production, and resultant cool-ing requirements, real estate needs, and so on). It may also free up your IT staff to be more productive — instead of spending all day managing hardware, they can focus on end-user needs.

Help Ease Power Consumption and “Stay Cool”

Virtualization provided the capability for multiple virtual machines with multiple operating environments to share a single AMD Opteron™ processor-based server. A range of applications, hosted under a variety of environments such as the Microsoft Windows, Linux, and Sun Solaris operating systems, can coexist on a single “real” machine. Each appli-cation executes within its own container, or virtual machine (VM), unaware that it might in fact be one of many applica-tions under one of many operating systems residing on a single piece of hardware.

Depending on the nature and duty cycle of the individually-hosted applications, a single AMD Opteron processor-based system can run hundreds of virtual machines. This results in a significant savings when considering the amount of power



Chapter 5: Ten Great Reasons to Invest in Virtualization Hardware 39it would take to run the same number of physical servers. These savings in real power consumption are amplified when you factor in the power needed (approximately 50 percent of required system power) to remove heat from the system. In addition to these forms of power savings, AMD is doing its part to reduce the power consumption within the processor, while delivering real-world performance and a full set of fea-tures. AMD Opteron processors come with the AMD-P suite of power management capabilities that dynamically minimizes power usage of the overall processor, individual cores, and the logic within each core, based on application needs and server workloads.

AMD-P enables the processor to provide the needed perfor-mance while reducing power consumption, which in turn produces less heat. Therefore, AMD-P technologies help control costs on two fronts: direct power consumption by the processors, and indirect power consumption required to deal with the heat thrown off by those processors.

Together, those technologies can help enable cutting-edge energy efficiency. That can be a major competitive advantage for cloud service providers, whose data centers typically house large volumes of power-hungry servers.

Provide Better SecurityVirtualization allows you to keep sensitive corporate data separate from end-user data, and one set of end-user data separate from another, even though they reside on the same physical machine. If you assign separate VMs to handle those data, you could give one VM with end-user tasks access to downloads and other higher-risk operations, while the VMs that handle other end-user data and sensitive and/or critical data stay securely out of reach.

Run Legacy Software on Non-Legacy Hardware

Many enterprises still have applications that will only run on older operating systems The last thing they want to do is try



Virtualization For Dummies, 2nd AMD Special Edition 40to keep old hardware around and useable just to run older applications.

A virtualized environment can enable a VM designed to emulate legacy hardware and allow you to run the necessary operating system and legacy software — all of this can run on the latest and greatest (and reliable) new hardware.

Develop and Test New Stuff Easily

Software developers can create virtual environments to run multiple operating systems, or even different versions of the same operating system, to test software.

Similarly, if you’re part of an enterprise considering rolling out new software, you can limit its deployment to one secure piece of your infrastructure (by installing it on a VM) and see how it behaves and how your users respond before you deploy it enterprise-wide.

Virtualization allows the rapid deployment of new VMs, along with the rapid relocation of VMs from one physical server to another. New applications can proceed from concept to development to deployment with fewer delays.

In addition, many applications that run on real (that is to say, nonvirtual) machines may require at least three dedicated systems: one for development, one for normal execution, and one to back up the system usually used for execution. Virtualization enables organizations to run the same work-loads on fewer physical systems, which can help the IT staff be more productive and more responsive to user needs.

Run Multiple Operating Systems on the Same Hardware

Many enterprises use multiple operating systems. It’s not uncommon to run Linux, Microsoft Windows, and Solaris



Chapter 5: Ten Great Reasons to Invest in Virtualization Hardware 41within one organization. Virtualization lets you do this more smoothly on the same hardware, rather than needing at least two discrete hardware installations.

Improve ScalabilityEnterprises grow. With virtualization, responding to the changing needs of your environment doesn’t need to follow the old way of “throwing more hardware at the problem.”

Instead, you can scale your environment much more dynami-cally, responding, for example, to growth in the user base of two server/application combinations with only one new piece of hardware virtualized to account for where you need the resources.

Enhance Your Hardware Utilization

With virtualization, you’re managing the aggregate demand of your enterprise instead of the demand on any particular server/application combination. You can deploy the appropri-ate total hardware complement to handle your peak aggregate loads rather than deploy individual pieces of hardware for every type of task. Doing so lets you keep your total hardware installation busy and earning its keep, instead of sitting idly by waiting for the next task that might suit a specific server/application combination.

Create a Manageable Upgrade Path

Upgrading software or operating systems can be a huge task. Virtualization lets you split up the task and do it at a pace slightly less insane than an all-nighter met in the morning by end-user panic. In a virtualized workplace, you can upgrade one group, spend some quality time getting to know them



Virtualization For Dummies, 2nd AMD Special Edition 42and how they’re getting along with their upgraded systems, and then use that knowledge to upgrade the next group on your list.

Manage Outages (Expected and Unexpected) Dynamically

For scheduled outages, a virtualized environment lets you shift resources to cover where the outage will occur. And with live migration tools, you can move virtual machines (and the users accessing applications on them) dynamically and transparently between systems. Your end-users need never skip a beat. Instead of getting an e-mail before the e-mail server goes down for routine maintenance, they never get an e-mail at all.

Virtualization can also help enhance disaster recovery: Should a VM fail for any reason, a new instance of a VM can be started at a remote location almost immediately.

Nice, huh?



Notes



Notes



Bernard GoldenMargaret LewisTim Mueting

Compliments of

• Create a dynamic data center

• Allocate memory where it’s needed

• Save energy, time, and money

• Improve security and scalability

Learn to:

Virtualization

2nd AMD Special EditionMaking Everything Easier!™

Open the book and find:

• A look at microprocessor advancements that help improve virtualization performance

• An overview of the different types of hypervisors

• How virtualization helps reduce power consumption in the data center

• An explanation of how memory is virtualized

Bernard Golden is a cloud computing expert and the author of Virtualization For Dummies. Margaret Lewis is currently director of Software Solutions for AMD, focusing on identifying the next gen software solutions for AMD products. Tim Mueting is responsible for AMD’s virtualization solution strategy.

ISBN: 978-0-470-64412-6 AMD tracking number: 42139-CNot for resale

Go to Dummies.com®

for videos, step-by-step photos, how-to articles, or to shop!

It seems like everywhere you go these days, someone is talking about virtualization. Technical magazines trumpet the technology on their covers. Virtualization sessions are featured prominently at technology conferences. Basically, virtualization is a technical innovation designed to increase the level of system abstraction and enable IT users to harness ever-increasing levels of computer performance.

• Virtualization’s background — understand where it came from and better know its moving parts

• Putting virtualization to work — what it can do for you and your company

• Hardware-assisted virtualization — AMD hard-ware features that help you get the most out of your virtualization environment

• Vir tualization for the desktop and client — server hosted desktops help improve manageability, security, and reduce costs

Understand how virtualization works and figure out whether it’s right for you

www.dummies.com