Dct Dummies

8/12/2019 Dct Dummies

1/52


2/52These materials are the copyright of John Wiley & Sons, Inc. and any

di i ti di t ib ti th i d i t i tl hibit d


3/52

Data CenterTransformation

Riverbed Special Edition

These materials are the copyright of John Wiley & Sons, Inc. and any



4/52

Data Center Transformation For Dummies, Riverbed Special Edition

Published byJohn Wiley & Sons, Inc.111 River StreetHoboken, NJ 07030-5774

www.wiley.com

Copyright 2014 by John Wiley & Sons, Inc.

Published by John Wiley & Sons, Inc., Hoboken, NJ

No part of this publication may be reproduced, stored in a retrieval system ortransmitted in any form or by any means, electronic, mechanical, photocopying,recording, scanning or otherwise, except as permitted under Sections 107 or 108 ofthe 1976 United States Copyright Act, without the prior written permission of thePublisher. Requests to the Publisher for permission should be addressed to thePermissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/

go/permissions.Trademarks:Wiley, the Wiley logo, For Dummies, the Dummies Man logo, AReference for the Rest of Us!, The Dummies Way, Dummies.com, Making EverythingEasier, and related trade dress are trademarks or registered trademarks of JohnWiley & Sons, Inc. and/or its affiliates in the United States and other countries, andmay not be used without written permission. Riverbed and the Riverbed logo areregistered trademarks of Riverbed. All other trademarks are the property of theirrespective owners. John Wiley & Sons, Inc., is not associated with any product orvendor mentioned in this book.

LIMIT OF LIABILITY/DISCLAIMER OF WARRANTY: THE PUBLISHER AND THE AUTHOR MAKENO 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

PROFESSIONAL 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 THEINFORMATION 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.

For general information on our other products and services, or how to create acustomFor Dummiesbook for your business or organization, please contact ourBusiness Development Department in the U.S. at 877-409-4177, contact [email protected], or visitwww.wiley.com/go/custompub. For information aboutlicensing theFor Dummiesbrand for products or services, contact BrandedRights&[email protected]

.ISBN 978-1-118-66064-5 (pbk); ISBN 978-1-118-66371-4 (ebk)

Manufactured in the United States of America

10 9 8 7 6 5 4 3 2 1




5/52

Table of Contents

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

About This Book ............................................................1

Icons Used in This Book ................................................2

Beyond the Book ............................................................2

Chapter 1: The Elements of Data CenterTransformation ............................................................. 3

People: Getting the Right IT Teamsto Work Together ........................................................4

Processes: Tackling Five Key Projects........................5

Technology: Choosing the Right Tools .......................6

Chapter 2: Making Applications Faster ................... 7

Seeing the Need for Application Speed .......................7

Using WAN Optimization to Accelerateand Prioritize Apps .....................................................8

Challenges .................................................................8

Improving Scalability and Speed withApp Delivery Technology .........................................10

Finding Tools That Monitor andTroubleshoot Applications ......................................11

Chapter 3: Converging Branch Infrastructure ...... 13

Employing WAN Optimization ....................................14Going One Step Further with Storage Delivery ........15

Virtualizing Branch Services ......................................15

Monitoring Branch Operations..................................16




6/52

Chapter 4: Protecting Branch Data ......................... 19

Centralizing All Your Data ..........................................20

Taking a New Approach to Storage Delivery ............20

Chapter 5: Consolidating Data Centers .................. 25

Recognizing the Key Aspects ofData Center Consolidation .......................................26

Using Performance Management Tools to

Discover Consolidation Requirements ...................26Redirecting Traffic with ADCs and

WAN Optimization ....................................................27

Selecting the right path withWAN optimization ................................................28

Redirecting users with ADC technology ..............29

Chapter 6: Managing Performance ........................ 31

Meeting Performance Management Challenges .......32

Relocating IT resources .........................................32

Diagnosing problems in complex networks ........33

Choosing Performance Management Tools..............33

Monitoring end-to-end ..........................................34Tracing application transactions .........................35

Troubleshooting performance issues ..................35

Creating customized dashboards ........................36

Integrating Performance Management Tools ...........37

Chapter 7: Ten Ways To Justify YourData Center Transformation .................................... 39

Improving User Productivity ......................................39

Delaying Bandwidth Upgrades ...................................40

iv




7/52

v

Eliminating Recreational Traffic ................................40

Reducing Server Infrastructure ..................................40

Centralizing Branch Infrastructure ............................41

Making Branch Backups Unnecessary ......................41

Using Cheaper Network Links ....................................41

Leveraging Cloud Storage ...........................................42

Improving IT Staff Productivity ..................................42

Consolidating Monitoring Tools................................42




8/52

Publishers Acknowledgments

Were proud of this book and of the people who worked on it. For

details on how to create a customFor Dummiesbook for yourbusiness or organization, contact [email protected] visitwww.wiley.com/go/custompub. For details on licensing theFor Dummies

brand for products or services, contact BrandedRights&[email protected].

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

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

Kathleen Nebenhaus, Vice President and Executive Publisher

Composition Services

Debbie Stailey,Director of Composition Services

Business Development

Lisa Coleman,Director, New Market and Brand Development

Acquisitions, Editorial,and Media Development

Project Editor:Carrie A. Burchfield

Editorial Manager:Rev Mengle

Acquisitions Editor:Kyle Looper

Business DevelopmentRepresentative:Karen Hattan

Custom Publishing ProjectSpecialist:Michael Sullivan

Composition Services

Sr. Project Coordinator:Kristie Rees

Layout and Graphics:Carrie A. Cesavice

Proofreader:Lindsay Amones


http://www.wiley.com/go/custompubhttp://www.wiley.com/go/custompubhttp://www.wiley.com/go/custompubhttp://www.wiley.com/go/custompubhttp://www.wiley.com/go/custompub


9/52

Introduction

Only a decade ago, consumers were happy if theycould look up such basic information as a stores

hours online. Today, users take the online presence ofeven the smallest businesses for granted, and they

want to access information fast often, from a mobiledevice.

The shift to empowered and savvy IT consumption isalso occurring in the corporate environment. Manyworkers feel that they have better technology at homethan they do at work, and nearly half are solving workproblems with personal technology.

Consumerization trends, such as Bring Your Own Device(BYOD), mobility, and cloud computing, pose new chal-lenges to IT departments worldwide. How does IT giveusers the better, faster, easier experience they wantwithout sacrificing the need for availability, scalability,and security? Meeting these challenges starts with

understanding how apps and data are delivered fromthe data center (enterprise-owned or in the cloud) andthen developing an improvement plan involving technol-ogies such as performance acceleration and performancemanagement.

About This BookThis book describes effective yet economical steps youcan take to improve user experience with applicationsdelivered from your data center. If youre wondering




10/52

2

how to make your users more productive, create morecost-effective data centers, and do both things without

replacing your existing infrastructure, this book is foryou.

This book is organized with multiple audiences in mind.Whether youre part of your IT departments networking,server, storage, or application team, you find relevantprojects here that help you deliver better businessresults.

Icons Used in This BookThroughout the book, special icons call attention toimportant information.

This icon points out information that may well

be worth committing to memory.

This icon marks especially technical materialthat you dont have to read (but I hope youwill).

This icon points out helpful suggestions andnuggets of useful information.

Warnings offer practical advice to help youavoid making potentially costly mistakes.

Beyond the BookFor more information aboutand resources on data

center transformation, visitwww.riverbed.com.


http://www.riverbed.com/http://www.riverbed.com/http://www.riverbed.com/


11/52

Chapter 1

The Elements of Data CenterTransformation

In This Chapter Getting the right IT roles involved

Focusing on five key projects

Putting the right technology in place

Change is never risk-free, especially in the IT world.The secret of success is using technologies that

provide a smooth transition to a higher-performance,more cost-effective infrastructure without a majorredesign.

Making such a transition requires coordinating theIT trifecta: people, processes, and technology. Thatmeans getting all the people involved on board, estab-lishing key projects that can achieve the desired goals,and figuring out which technology solutions to use.

This chapter is for you if

Youre a CIO or VP of infrastructure whos lookingat new ways to improve your IT performance with-out breaking the bank




12/52

4

Youre a network, storage, or application managerwho wants to improve collaboration with other

parts of your IT organization

People: Getting the Right ITTeams to Work TogetherIn the IT-centric world of only a few years ago, each

domain (such as storage, networking, applications, andserver) was managed by a team of professionals whowere dedicated solely to that particular part of infra-structure. Each team operated in its own silo. Anychange was very slow, and the idea of asking userswhat they wanted rarely crossed anybodys mind.

Today, IT is no longer just a cost center, but also a

driver of innovation for a business. In this user-centricera, all pieces of the IT puzzle must work togetherperfectly.

A successful data center transformation should makecollaboration easier and should help each team achieveits goals. That means focusing these five roles on

Network:Reducing bandwidth cost, automatingperformance monitoring, and improving appresponse times

Storage:Securing data, ensuring that data isalways accessible, and decreasing storage andbackup costs

Server:Virtualizing and consolidating resources,

making those resources elastic, and providingglobal load balancing across data centers




13/52

5

Application development:Accelerating therelease of new apps, decreasing the cost of app

production, and improving communication withoperations teams

Application operations:Increasing the availabilityand visibility of application performance, ensuringapplication scalability, and reducing troubleshoot-ing time

Processes: TacklingFive Key ProjectsOrganizations that make successful large-scale improve-ments in their data centers focus on the following fiveprojects:

Making applications faster: Providing great per-formance for all users everywhere

Converging branch infrastructure: Improving localperformance without adding local complexity

Protecting branch data: Enabling users to accessdata anywhere without that data leaving the data

center Consolidating data centers: Giving users the

same experience or better with fewer data centers

Managing performance: Monitoring, troubleshoot-ing, and accelerating application performance

I cover all these projects in detail in Chapters 26.




14/52

6

Technology: Choosing

the Right ToolsThe key enablers of data center transformation fall intotwo categories: performance acceleration and perfor-mance management. A comprehensive application per-formance acceleration solution makes both users andIT more productive. Users can accomplish more in a

shorter time when applications are available at greatspeeds at any time and on any device. Also, IT canspend less time troubleshooting and more time drivinginnovation for business if it has efficient performancemanagement tools.

For details on specific tools that can help with datacenter transformation, see Chapters 26.




15/52

Chapter 2

Making Applications Faster

In This Chapter Accelerating and prioritizing your applications

Improving application scalability and speed

Monitoring application performance

A

re you responsible for application performance?Would you like to reduce user complaints aboutslow applications? If you answered Yes to both ques-tions, this chapter is for you.

Seeing the Need for

Application SpeedYour business needs applications of all kinds collaboration, e-commerce, business suites, and soon that are rapidly available to both internal andexternal users. Applications let businesses eliminatemanual, paper-based processes; reduce employeetravel; and lower other costs of doing business.

As applications become more complex, performancebottlenecks may occur at several points, such as theservers running the application, the network delivering




16/52

8

the application, or the application rendering the con-tent. To improve user experience, you can do one or

more of the following things: Accelerate delivery of applications to remote

users, who often experience slow uploads anddownloads, page loads, and other functions

Prioritize applications based on business needand service level agreements

Scale applications to ensure that applicationscontinue to deliver high performance as user andbusiness needs grow

Monitor applications, servers, and network infra-structure to detect performance problems andpinpoint their causes, ultimately acceleratingproblem resolution

I discuss all these strategies in this chapter.

Using WAN Optimization toAccelerate and Prioritize Apps

Delivering slow applications to your remote employeescan put your business at a distinct competitive disad-vantage and decrease user satisfaction. Those usersdepend on the network for their applications, as do allemployees who access applications in the cloud.Speeding network traffic all the way out to users, wher-ever they may be, is clearly a priority, but multiple fac-tors can slow your mission-critical applications.

ChallengesIts challenging to provide users on the fringes of anetwork the same performance that theyd receive at




17/52

9

headquarters. Communications speeds over the net-work are limited by available bandwidth, latency, and

packet loss.

Packet loss and network congestion are net-work characteristics that severely affect thespeed of applications and data transfer.

WAN optimization uses several techniques in a singlecontroller, so you can holistically overcome perfor-

mance impediments. Five techniques within a WANoptimization solution are

Data streamlining:Eliminates redundant datatransfers and, as a result, reduces WAN band-width use by up to 95 percent

Transport streamlining:Eliminates transportprotocol inefficiencies and reduces the number

of TCP packets required to transfer data by up to98 percent

Application streamlining:Provides additionaloptimizations for application protocols

Also, by reducing application protocol chattinessby up to 98 percent and minimizing application

overhead, application streamlining increasesthroughput.

Quality of service (QoS):Allows IT to identify andprioritize business-critical applications to ensureoptimal user experience for real-time and interac-tive applications

QoS also helps restrict non-business-critical appli-cations that consume significant network traffic.

Path selection:Like QoS, allows IT to send priori-tized traffic across the most appropriate networkconnection




18/52

10

IT can deliver applications on the network linkthat provides the best performance, cost, andreliability characteristics.

Improving Scalability and Speedwith App Delivery TechnologySometimes, applications are slow for reasons beyondnetwork problems. Your applications may be slow orfailing because theyre overwhelmed by the volume ofusers or the amount of content on a web applicationpage.

Its difficult to predict how many users are going toaccess an application at the same time and make sure

Candidate apps for WAN optimizationApplications that users access over a WAN and/or the cloudare candidates for WAN optimization, including the following:

Collaboration and unified communications applications,including document sharing, voice, and video applications

Business-suite applications, such as enterprise resourcemanagement and customer relationship managementapplications

Application and desktop virtualization, such as virtualdesktop infrastructure (VDI), terminal services, and appli-cation streaming software

Software as a Service (SaaS) applications used byemployees




19/52

11

that they all get a fast application experience. Applicationavailability and speed are limited by the number of server

connections, server processing resources, bandwidth,and latency.

You may need an application delivery controller (ADC).ADCs are advanced load-balancers, designed to makeapplications run faster, more reliably, and more securelyby offloading some of the work that web servers wouldotherwise have to perform. This strategy boosts the

throughput of the web-server tier in an application. Inaddition to load balancing application workloads acrossvarious environments, ADCs optimize user services byinspecting, transforming, prioritizing, and routing appli-cation traffic.

Finding Tools That Monitor andTroubleshoot ApplicationsAs applications have become more critical, theyvealso become more complex. Infrastructure changes,application code updates, and mobile user devices andnetworks constantly introduce new variables to appli-

cation execution and delivery. In the face of ongoingchange, you need a continuous, end-to-end view ofapplication and infrastructure performance.

Figuring out why an application is performing slowly ornot responding at all can be like finding a needle in ahaystack. Resolving performance issues quickly becomesextra-complicated with cloud IT services, service-oriented

architectures (SOA), virtualization, and mobility.A complex application infrastructure requires a perfor-mance management solution that looks across the




20/52

12

application delivery chain. Whether youre rolling outnew applications, consolidating or virtualizing data

centers, or migrating to the cloud, managing applica-tion performance requires a holistic view that takesinto account user experience, transaction tracing, com-ponent monitoring, and infrastructure and networkperformance management.




21/52

Chapter 3

Converging BranchInfrastructure

In This Chapter Centralizing branch resources

Virtualizing and consolidating branch services

Monitoring branch IT operations

As branch-office employees take more responsibilityfor sales and customer relationships, maintaining

productivity and reducing business disruption becomemore critical. Any interruption of service negativelyaffects productivity in the branch, which negatively

affects the business as a whole.For IT, branch offices have traditionally been hard tomanage and costly to maintain. Centralizing servers,storage, and even desktops from branch locations intocentral data centers reduces the complexity of branchIT infrastructure and allows a business to take advan-tage of data center cost efficiencies. IT must ensure,




22/52

14

however, that it can converge the branch operations toa small infrastructure footprint without compromising

user performance and still provide efficient manage-ment with maximum availability of services.

If youre responsible for storage and data backup, or ifyou want to consolidate and manage data and applica-tions in the data center while making them accessibleat branch offices at local speeds, youve come to theright place. This chapter is for you.

Employing WAN OptimizationMany organizations choose to centralize branch officeapplications and data to reduce costs and complexity.Providing LAN-like performance and efficiency forremote users is critical to the success of branch con-

vergence, yet most applications arent designed tooperate over a network. Also, bandwidth limitationsand protocol inefficiencies are exacerbated by distanceand latency.

Many organizations have turned to WAN optimizationto address bandwidth and latency concerns. WANoptimization products combine data reduction, trans-

port protocol optimization, and application protocoloptimization, providing WAN application accelerationand bandwidth reduction to centralized applicationsthat users in branch offices can access. This techniqueeliminates the need to upgrade network bandwidth andis one of the most significant benefits of deployingWAN optimization appliances.




23/52

15

Going One Step Further

with Storage DeliveryWAN optimization is necessary but not always sufficient.You can fully centralize all your branch resources inyour data center without also tackling the need forLAN-like performance for storage resources by using anemerging technology called storage delivery.

Storage delivery works similarly to WAN optimization.It uses a series of streamlining techniques to optimizethe delivery of data across networks, thereby overcom-ing the distance, latency, and packet-loss characteris-tics that can plague networks. The result is that yourbranch users can access file servers, application serv-ers, and databases now located thousands of miles

away in your data center.WAN optimization also allows users to take advantage ofstorage-intensive technologies such as virtual desktopinfrastructure (VDI). Traditionally, VDI hasnt workedwell on WANs, but storage delivery eliminates the per-formance issues that lead to poor user experience.

I discuss the specifics of storage delivery (andshow how it also increases data security) inChapter 4.

Virtualizing Branch ServicesConsolidation and virtualization are rapidly changingthe way that organizations manage their branch-office




24/52

16

IT infrastructures, providing greater flexibility, improv-ing service delivery, and reducing costs. Many of these

benefits can be extended to local branch services suchas print, dynamic naming service (DNS), and dynamichost configuration protocol (DHCP).

When you virtualize branch services on a single, con-verged appliance, you lighten the IT footprint by pro-viding a simplified, consolidated architecture thatstreamlines operations, simplifies management, and

enables agile service delivery. You can also savemoney. One organization saved $10,000 to $15,000 inequipment costs in its new offices by running localbranch services on a converged appliance.

Monitoring Branch OperationsMany IT administrators rely on user complaints todetect performance issues. But advanced integratedmanagement tools allow you to monitor and trouble-shoot any issues that arise before they become prob-lems. By automatically mapping IT environmentsnetworkwide, these tools speed consolidation effortsand give the organization a current, accurate view of

infrastructure inventory.These capabilities are provided by network performancemanagement (NPM) and application performance man-agement (APM) tools, which I cover in Chapter 6.




25/52

17

Develop a plan for resiliencyConsolidating branches delivers compelling savings, but ITleaders need to carefully consider the resiliency of WANarchitecture to ensure uptime and continuity for critical busi-ness applications and functions. WAN optimization technol-ogy can help. One global leader in business technologydeployed WAN optimization technology and reduced its

backup windows from 24 to 6 hours, decreased bandwidthuse by 90 percent, and ensured data was kept securely in thedata center.




26/52

18




27/52

Chapter 4

Protecting Branch Data

In This Chapter Centralizing storage in the data center

Protecting data without sacrificing performance

Seeing how storage delivery technology works

W

hen it comes to data, IT has two goals: to makeit accessible and responsive so users can use itproductively and to keep it safe. In the past, achievingboth goals simultaneously in branch offices has beendifficult. Data is safe when its stored at the data center,where skilled IT professionals back it up and guard itbehind a robust perimeter. When data is left anywhereoutside the data center, however, its vulnerable andexposed to risks.

If youre responsible for branch infrastructure or appli-cations, or if you seek to consolidate and manage allyour data in the data center while making it accessibleat branch offices at local speeds, keep reading. Thischapter is for you.




28/52

20

Centralizing All Your DataThe same storage delivery technology that I discuss inChapter 3 can be applied to data security so that youno longer have to compromise data safety for accessi-bility. Data centralization also eliminates the hassleand cost of performing local backups in the branches.

The latest innovation in storage delivery allows you toconsolidate and manage all your data in the data centerwhile making it accessible to branch offices at localspeeds. In this approach, storage is decoupled from itsserver but actually works as though it were local. Auser in Paris, for example, can access a virtual server inthat city, which in turn accesses data in Chicago, butthe users experience is local.

Taking a New Approachto Storage DeliveryLow-level storage protocols like SCSI are designed towork over short distances (with low latency and highbandwidth) via reliable media. SCSI over TCP/IP is

deployed only inside a data center, where distancesare limited, connectivity issues are absent, and band-width is ample. Now, however, the SCSI protocol canwork over long distances with limited bandwidth overunreliable media. This new approach to storage deliv-ery requires the installation of two components:




29/52

21

Data-center component:The first element isinstalled in your data center and connects to the

storage system of your preferred vendor, usingindustry-standard low-level storage protocols.Then it projects the storage to the branch.

The data-center component also inspectsmounted file systems and proactively streamsdata to the branch locations via block-level pre-diction algorithms (key software that predicts and

fetches blocks needed at branch offices). Thiscapability allows virtual servers and data fromcentralized storage to be available wherever andwhenever theyre needed.

Branch component:The second componentresides in the branch office itself. This branch(sometimes referred to as edge) component com-municates with its data center counterpart andpresents the remote data to the branch for use bylocal servers and applications.

In addition, the branch component caches work-ing data to deliver LAN-like performance tobranch users and applications. It also extends

thin provisioning to the remote site, because notall data is required to work in the branch. Finally,the branch component supports fast cold writes,which essentially means that applications getfast local access and data is streamed back tothe data center.




30/52

22

Can storage delivery help you?These questions may determine whether storage deliverycan help protect your organizations data:

Does your organization still host servers and/or storagein branches?By consolidating data in the data center, astorage delivery system improves data safety and allowsyou to simplify your remote operations.

How is data protected in branch offices?Outside thedata center, your data is vulnerable to multiple risks. Astorage delivery system keeps data in a secure datacenter and encrypts that data both while its in use at thebranches and while its in transit to and from the datacenter.

Does your organization perform remote backups?Bycentralizing your data, a storage delivery system elimi-nates the hassle and cost associated with branchbackup, saving you 50 percent on average.

How quickly can your branches recover from disaster?With the storage delivery approach, you can reconnectto the data center and boot branch servers back into ser-

vice to resume operations in minutes. How quickly can you provision an entire branch?With

new ways of delivering storage, you can provision cen-trally, connect, and start branch services in new loca-tions within minutes.




31/52

23

Dont be confused by similar-soundingapproaches. Proper storage delivery solu-

tions must operate at block level to remainapplication- and data-agnostic. Object-levelcaching solutions arent the same as the stor-age delivery approach I describe here. Thesesolutions fail to achieve the data performanceand integrity of block-level storage deliverysolutions.




32/52

24




33/52

Chapter 5

Consolidating Data Centers

In This Chapter Using planning and discovery tools

Redirecting users and traffic

Considering virtual technologies

M

ultiple data centers have become the norm asorganizations expand their operations. For ITdepartments, this trend has led to more costs andbigger infrastructure management headaches. To com-pete in a highly competitive global market, businessesmust find a way to reduce the number of data centerswithout negatively affecting user experience.

Reducing the number of data centers in an enterprisecan produce significant reductions in cost and com-plexity. At the same time, without the need to operateredundant data centers, IT becomes more efficient andcan scale resources to deliver services to branches andusers effectively.




34/52

26

This chapter is for you if

Youre responsible for data center infrastructureand operations.

Youre in charge of storage and data backup.

You want to operate fewer data centers, achieveconsistent high performance across a distributedorganization, and streamline administration anddata center protection.

Recognizing the Key Aspectsof Data Center ConsolidationThe right approach to consolidating data centersenables IT departments to reduce cost and streamline

administration tasks with no effect on applicationperformance.

A successful data center consolidation project has fourkey aspects: acceleration, scaling, discovery, and redi-rection. I discuss acceleration and scaling in Chapter 2as part of improving application performance. In thefollowing section, I elaborate on the two other require-

ments: discovery and redirection.

Using Performance ManagementTools to Discover ConsolidationRequirementsProper planning and discovery ensure minimal impactfrom a consolidation project. Dependency mapping, for




35/52

27

example, should be part of preparing for a consolida-tion project, because even one change may affect every

user on the network. Application and network monitor-ing tools provide visibility into virtualized environ-ments and traffic performance. This visibility enablesIT organizations to map global IT environments andreduce the risk of data center migrations and networkoutages by highlighting application dependencies.

You can accomplish this dependency mapping with the

right performance management tools. These toolsgreatly improve IT productivity and efficiency, as well asensure better application performance. IT organizationscan use these tools to identify and diagnose perfor-mance problems faster, thereby reducing recovery time.I talk about these tools in more depth in Chapter 6.

Look for performance management tools thatprovide visibility and actionable insightinto network and application performance.These tools should offer a combination of user-experience monitoring, application-transactiontracing, and network and IT infrastructure per-formance management.

Redirecting Traffic with ADCsand WAN Optimization

As you consolidate to fewer data centers, you have to besmart about where you send traffic. Do you send a userto the closest data center for her application? What if

that data center is offline or at capacity already? Youneed technology that can help you answer these ques-tions and automatically redirect traffic as necessary.




36/52

28

WAN optimization and ADC technologies not only accel-erate performance for critical applications, but also

ensure efficient load balancing of users across datacenters based on performance, availability, and costconsiderations to deliver the best user experience.

Selecting the right pathwith WAN optimizationThe right WAN optimization technology provides LAN-like access to data and applications across a WAN any time, anywhere throughout the enterprise formobile workers, branch offices, and data centers. Italso facilitates management of complex hybrid net-works by using path selection, described in Chapter 2.

A hybrid network is one that consists of multiple net-

work link types, such as multi-protocol label switching(MPLS), IP virtual private networks (VPN), and com-modity Internet connections. Path selection allows youto prioritize which traffic traverses which network link.You can send traffic out on a specific link based on per-formance, cost, security, or availability criteria.

Traditionally, companies backhauled Internet traffic

across a private network, sent the data off to theInternet, and then routed it back across the private net-work to the end user. These secure links are often quiteexpensive, so the economics of this approach are poor.Also, the user experience is terrible. Because so muchInternet traffic now consists of mission-critical applica-tions, this double trip of Internet traffic just clogs the

central network. A direct-to-net link would be fasterand more cost-effective.




37/52

29

Redirecting traffic across hybrid networks isessential in highly consolidated environments.

Without this redirection you risk providing apoor user experiencing and incurring highernetwork costs.

Redirecting users with ADC technologyConsolidating data centers with ADC technology inplace means that users dont feel the effects of migrat-

ing data centers or moving applications. The ADC tech-nology offers a single point of control for all users andcan redirect those users locally or globally.

Local redirectionpoints a user to the best resourcewithin the data center. This technique may mean send-ing a user to a server or virtual machine thats usedless than the others or that has specific computing

capabilities.

Global redirectionpoints a user to a different datacenter one that has more capacity, higher perfor-mance infrastructure, or greater resilience.

Look for ADCs that offer advanced featuressuch as web content optimization and applica-

tion firewall capability. Software ADCs giveyou more flexibility and scalability with off-the-shelf hardware. Also look for industry-standard application programming interfaces(APIs) like REST so that you can program yourADC, as well as integrate it with technologiessuch as performance management tools.




38/52

30

Is a virtual ADC right for you?If your answer to any of the following questions is yes, yourorganization may benefit from a software-based virtual ADC:

Are you virtualizing the applications that youre load-balancing?

Do you need to scale ADC capacity up and down quickly?

Are you running an application or part of an applicationin the cloud (such as Microsoft Windows Azure, Joyent,or Rackspace)?

Are you trying to shorten application version releasecycles?




39/52

Chapter 6

Managing Performance

In This Chapter Understanding the challenges

Choosing the right tools

Integrating your performance management suite

M

anaging performance is an everyday concern forpeople who are charged with maintaining appli-cation and data center infrastructure, but its evenmore critical for an IT department thats involved in alarge-scale data center transformation effort. Such anIT department requires sophisticated monitoring tools.The consolidation of multiple data centers provideseven more opportunity for performance problems toarise, and any kind of structural change requires IT topay extra attention to user experience, trace transac-tions, and monitor network performance. Powerful,easy-to-use performance management tools are thekeys to maintaining quality of service.




40/52

32

Youll likely want to read this chapter if youre

Responsible for application support

Struggling to identify, troubleshoot, and resolveapplication performance issues

Responsible for successfully migrating applica-tions during data center consolidation orrelocations

Meeting PerformanceManagement Challenges

To take advantage of all that modern performance man-agement tools have to offer, IT operations and applica-tion teams need to overcome two critical challenges:

relocating IT resources and diagnosing problems incomplex networks.

Relocating IT resourcesThe first challenge is relocating applications, whetherrelocation is due to branch consolidation, data centerconsolidation, or the use of public cloud services. IT

organizations then need the flexibility to relocate otherIT resources, such as server and storage infrastructure.Traditional tools perform manual discovery and map-ping, but that process is slow and error-prone. Agent-and scan-based discovery and mapping tools areexpensive, resource-intensive, and potentially limitedin scope.




41/52

33

Diagnosing problems in complex networks

Virtualization is blurring the traditional lines of server,storage, and network infrastructures. Users expectapplications to be fast and available, but redundantnetworks, interdependent application tiers, and virtual-ized servers and networks make diagnosing an applica-tion performance issue extremely complex.

Legacy tools provide only a fragmented view of opera-

tions. Also, IT operations are overloaded with vastamounts of performance data but cant derive anyactionable information from it. As a result, problemidentification, diagnosis, and resolution can take daysor weeks.

Choosing PerformanceManagement Tools

Performance management tools can help your organi-zation quickly identify, troubleshoot, and resolveperformance issues, as well as plan and optimize infra-structure deployments.

Your organization may already have dozens (if not hun-dreds) of performance management point tools. Someof these tools are specific to network, server, or stor-age; others provide only monitoring, dependency map-ping, or analysis capabilities. To manage performanceeffectively across a modern enterprise, however, youneed a performance management suite.




42/52

34

Focus on a suite that enables you to do fourthings: monitor, trace, troubleshoot, and cus-

tomize. I discuss all four tasks in the followingsections.

Monitoring end-to-endProblems can occur anywhere, from a device to thedata center. To diagnose performance issues accu-rately, you need to be able to monitor all aspects of

the environment.Monitoring starts with users, because their experienceis the ultimate criterion by which a system is judged. Ifperformance starts to degrade for users, its better foryou to know before users even notice that things havestarted to slow down.

You can increase user productivity by 7 percentby proactively fixing user problems.

End-to-end user visibility requires the ability to moni-tor whats happening on an actual device, as well asfine-grained packet and flow level monitoring to moni-tor user response times.

In addition, you need comprehensive infrastructuremonitoring tools, which focus on the network.Comprehensive infrastructure monitoring tools notonly go deep on network monitoring, but also provideinsight into the rest of the data center infrastructure.

Make sure that your monitoring tools areapplication-aware, capable of analyzing spe-

cific protocols and traffic types and thenreporting in terms of the actual applicationsinvolved.




43/52

35

Tracing application transactions

Application transaction tracing helps you track, moni-tor, and troubleshoot the performance of all compo-nents of multiple-tier applications and the transactionsthat link them. Following transactions complementsyour monitoring efforts, giving you a true, 360-degreeview of performance in your environment.

In addition, you can use these tools to get

accurate application-dependency maps fantastic tools to use before you begin anymigration activities in your data center trans-formation efforts.

Troubleshooting performance issuesYou can decrease outages by 67 percent by trouble-

shooting application performance issues. But to ensurethat you can uncover all types of performance issues,you need detailed analytics.

Good troubleshooting tools offer the following features:

Offline, multiple-tier analysis of a singletransaction

Transaction delay breakdown Application capacity planning

What-if scenario modeling

To get broad coverage, you need troubleshoot-ing tools that integrate with a wide variety ofmonitoring tools. A tool that incorporates flow,

packet, user, and transaction trace data pro-vides the best troubleshooting analytics.




44/52

36

You also want tools that offer try this rec-ommendations and transaction modeling

to provide proactive or even predictivetroubleshooting.

Creating customized dashboardsThe final task in a good performance managementstrategy is creating customized dashboards, becausevisualizing application and user performance is critical

to IT and to business executives.From an IT perspective, a well-crafted dashboardensures that you can make quick decisions on the over-all health of your IT environment. As part of your datacenter transformation, you want to track applications,locations, users, and infrastructure types to make surethat things are progressing smoothly. A comprehensive

performance management suite provides this level oftracking and allows you to transition quickly from view-ing data to taking action.

From a business perspective, dashboards ensure thatlarge projects (such as data center transformation) areoccurring on time and on budget. An executive dash-board may be simpler than an IT dashboard, highlight-ing the health of specific applications broken down byrelevant lines of business. The dashboard also maypresent relevant data about any chargeback modelsyouve implemented.




45/52

37

Integrating Performance

Management ToolsTo take better advantage of your performance manage-ment suite, integrate it with the rest of your data centertransformation investments. WAN optimization andapplication delivery controllers (ADCs), for example,can serve as data-capture devices that feed into your

performance management environment. These control-lers can reduce the cost of deploying dedicated flowand packet-capture devices.

You can go one step further by creating performanceautomation loops. Your WAN optimization controller,for example, is capable of shaping and prioritizingtraffic. If you integrate performance management with

WAN optimization, you can trigger automatic actionsbased on performance problems as they arise. Similarly,you can measure the user experience of a particular webapplication and then automatically trigger a content-optimization or load-balancing action on your ADC.




46/52

38




47/52

Chapter 7

Ten Ways To Justify Your DataCenter Transformation

In This Chapter Delaying costly upgrades

Eliminating unnecessary infrastructure

Saving IT personnel time and energy

Transforming your data center is the key to meetingchanging business requirements and rising user

expectations. If your organization is like most, though,you need to justify your IT projects.

The good news is that data center transformationefforts provide compelling benefits. This chapter out-lines ten areas in which performance acceleration andperformance management technologies can benefitvirtually any organization.

Improving User ProductivityArguably, ITs most important mission is to keepemployees productive and happy. Most employees,




48/52

40

however, spend minutes or even hours accessing slowapplications and data. Performance acceleration tech-

nologies such as WAN optimization, application deliv-ery, and storage delivery remove this obstacle. Keepingusers productive not only provides direct, immediatesavings, but also saves money over the long term bybolstering employee satisfaction and retention.

Delaying Bandwidth UpgradesIn addition to accelerating applications, WAN optimiza-tion typically reduces bandwidth consumption by6095 percent. That savings can enable you to delay abandwidth upgrade project, which can cost thousandsof dollars per site or tens of thousands of dollars forinternational sites in remote and emerging nations.

Eliminating Recreational TrafficChances are that video traffic from websites such asYouTube or Netflix consumes up to 50 percent of yourcorporate bandwidth. Some of this traffic may bebusiness-critical, but most of it is likely to be recre-ational traffic. Quality of service (QoS) can prioritizebusiness-critical applications and restrict network useto sites that drive business value, which can alsoreduce bandwidth needs and allow you to postponeupgrades (see the preceding section).

Reducing Server InfrastructureApplication delivery controllers (ADCs) improve theuptime of an application and reduce the number of webservers needed for an application by making serversmore efficient. As a result, you can further consolidate




49/52

41

and virtualize your server infrastructure, as well aseliminate unnecessary costs.

Centralizing BranchInfrastructure

Combine WAN optimization and storage delivery to cen-tralize servers, applications, and data from your remote

sites, all without affecting user experience. Centralizingyour branch infrastructure eliminates the need to rundedicated servers and storage at remote sites and cancut IT costs at each location by 3050 percent.

Making Branch Backups

UnnecessaryStorage delivery technologies eliminate storage at yourbranch locations, which has many ancillary benefits.You no longer have to run expensive, dedicated branchbackup solutions, which often require software, backupmedia such as tape, and local personnel. You can multi-ply these cost savings by the number of sites where

you use storage delivery.

Using Cheaper Network LinksBuilding hybrid networks (as described in Chapter 5)can dramatically change the economics of your net-work. Consider using cheaper Internet links and virtual

private networks (VPNs) to get as much as five timesthe bandwidth for one fourth the cost of traditionalWAN choices.




50/52

42

Leveraging Cloud StorageBacking up all your data center storage can be costly.Even if youve moved to technologies such as virtualtape libraries, you still need to maintain in-house infra-structure to store backup and archive data. Cloud stor-age of data, however, costs only a few pennies pergigabyte per month. Many companies are eliminatingexpensive backup and archive infrastructure. Instead,

theyre saving up to 80 percent by using storage deliv-ery technologies that leverage public cloud storage.

Improving IT Staff ProductivityPerformance management solutions that provide anend-to-end view of users, applications, and infrastruc-

ture can improve IT efficiency by enabling faster prob-lem identification, quicker mean time to resolution, andfewer on-site visits to resolve performance issues.

Consolidating Monitoring ToolsA comprehensive performance management suite

reduces the need for separate monitoring tools to trackuser experience; trace application transactions; andmonitor components, servers, the network, and otherinfrastructure. Many organizations can eliminate adozen or more separate products by using a perfor-mance management suite.




51/52These materials are the copyright of John Wiley & Sons, Inc. and any



52/52

Dct Dummies

Documents

Transcript of Dct Dummies