VCA-DCV Official Cert Guide

From the Library of David Alfaro Cordova
VCA-DCV
Official Cert Guide

VMware Press is the official publisher of VMware books and training materials, which
provide guidance on the critical topics facing today’s technology professionals and
students. Enterprises, as well as small- and medium-sized organizations, adopt
virtualization as a more agile way of scaling IT to meet business needs. VMware Press
provides proven, technically accurate information that will help them meet their goals for
customizing, building, and maintaining their virtual environment.
With books, certification and study guides, video training, and learning tools produced
by world-class architects and IT experts, VMware Press helps IT professionals master a
diverse range of topics on virtualization and cloud computing. It is the official source of
reference materials for preparing for the VMware Certified Professional examinations.
VMware Press is also pleased to have localization partners that can publish its prod-
ucts into more than 42 languages, including: Chinese (Simplified), Chinese (Traditional),
French, German, Greek, Hindi, Japanese, Korean, Polish, Russian, and Spanish.
For more information about VMware Press, please visit vmwarepress.com.

pearsonitcertification.com/vmwarepress
Complete list of products • Podcasts • Articles • Newsletters
VMware® Press is a publishing alliance between Pearson and VMware,

and is the official publisher of VMware books and training materials that provide
guidance for the critical topics facing today’s technology professionals and students.
With books, certification and study guides, video training, and learning
tools produced by world-class architects and IT experts, VMware Press
helps IT professionals master a diverse range of topics on virtualization
and cloud computing, and is the official source of reference materials
for completing the VMware certification exams.
Make sure to connect with us!

informit.com/socialconnect

This page intentionally left blank

VCA-DCV
Official Cert Guide
Matthew Vandenbeld
Jonathan McDonald
Upper Saddle River, NJ • Boston • Indianapolis • San Francisco

New York • Toronto • Montreal • London • Munich • Paris • Madrid
Capetown • Sydney • Tokyo • Singapore • Mexico City

VCA-DCV Official Cert Guide ASSOCIATE PUBLISHER
David Dusthimer
Copyright © 2014 VMware, Inc.
ACQUISITIONS EDITOR
Published by Pearson plc Joan Murray
Publishing as VMware Press VMWARE PRESS
All rights reserved. Printed in the United States of America. This publication is PROGRAM MANAGERS
protected by copyright, and permission must be obtained from the publisher prior Anand Sundaram
to any prohibited reproduction, storage in a retrieval system, or transmission in DEVELOPMENT EDITOR
any form or by any means, electronic, mechanical, photocopying, recording, or Ellie Bru
likewise. MANAGING EDITOR
ISBN-13: 978-0-13-385449-7 Sandra Schroeder
ISBN-10: 0-13-385449-3 PROJECT EDITOR
Seth Kerney
Text printed in the United States at Courier in Westford, Massachusetts.
COPY EDITOR
First Edition May 2014 with corrections June 2014
Megan Wade-Taxter
All terms mentioned in this book that are known to be trademarks or service
PROOFREADER
marks have been appropriately capitalized. The publisher cannot attest to the
Sarah Kearns
accuracy of this information. Use of a term in this book should not be regarded as
affecting the validity of any trademark or service mark. COORDINATOR
Vanessa Evans
VMware terms are trademarks or registered trademarks of VMware in the United
States, other countries, or both. BOOK DESIGNER
Chuti Praserstith
Warning and Disclaimer COMPOSITION
Every effort has been made to make this book as complete and as accurate as pos- Bronkella Publishing
sible, but no warranty or fitness is implied. The information provided is on an
“as is” basis. The authors, VMware Press, VMware, and the publisher shall have
neither liability nor responsibility to any person or entity with respect to any loss
or damages arising from the information contained in this book or from the use
of the DVD or programs accompanying it.
The opinions expressed in this book belong to the author and are not necessarily
those of VMware.
Special Sales
For information about buying this title in bulk quantities, or for special sales
opportunities (which may include electronic versions; custom cover designs; and
content particular to your business, training goals, marketing focus, or branding
interests), please contact our corporate sales department at corpsales@pearsoned.
com or (800) 382-3419.
For government sales inquiries, please contact governmentsales@pearsoned.com.
For questions about sales outside the U.S., please contact international@pear-
soned.com.

Contents at a Glance
Introduction xviii
CHAPTER 1 Datacenter Virtualization 3
CHAPTER 2 Business Challenges 21
CHAPTER 3 vSphere Components 41
CHAPTER 4 vSphere Storage 77
CHAPTER 5 VMware vSphere Networking 101
CHAPTER 6 Virtualization Solutions That Solve Business Challenges 125
APPENDIX A Answers to the “Do I Know This Already?” Quizzes and Chapter Review
Questions 143
Glossary 153

Table of Contents
About the Authors xiii
Dedications xiv
Acknowledgments xv
About the Reviewers xvi
We Want to Hear from You! xvii
Reader Services xvii
Introduction xviii
Chapter 1 Datacenter Virtualization 3
“Do I Know This Already?” Quiz 3
Foundation Topics 6
What Is Virtualization? 6
Datacenter Components 7
Compute 8
Storage 9
Why Virtualize? 11
Optimization 11
Availability 12
Management 13
Scalability 14
Summary 15
Exam Preparation Tasks 16
Review All the Key Topics 16
Definitions of Key Terms 17
Chapter Review Questions 17
Chapter 2 Business Challenges 21

Foundation Topics 24
Availability Challenges 24
What Is Availability? 24
Common Business Challenges 25
Hardware Availability 25
Application Availability 26
Maintenance 26
Environmental Availability 26
Management Challenges 27
What Is Management? 27
Datacenter Management 28
Provisioning Management 29
Compliance Management 29

Table of Contents ix
Scalability Challenges 29
What Is Scalability? 29
Infrastructure Scalability 31
Virtual Machine and Application Scalability 31
Optimization Challenges 32
What Is Optimization? 32
Monitoring Optimization 33
Efficiency Optimization 33
Performance Optimization 34
SMB Versus Enterprise Challenges 34
Size of the Company 34
Summary 36
Review Questions 37
Chapter 3 vSphere Components 41

Virtualization Concepts 45
Virtual Resources 45
Processor 46
Memory 46
Hard Disk 47
Network 47
Virtual Machines 48
Hypervisors 50
ESXi Hosts 50
Others 51
Managing Resources 51
vCenter Server 52
Clusters 53
Managed Hosts 53
Clusters of Hosts 53
Resource Pools 54
Distributed Resource Scheduler 55
DRS Scheduling Functions 55
Distributed Power Management Functions 56

x VCA-DCV Official Cert Guide
High Availability and Disaster Recovery 57

VMware High Availability 57
VMware Fault Tolerance 59
VMware Application High Availability 60
vSphere Data Protection 61
VMware vCenter Site Recovery Manager 62
vSphere Replication 64
Migration 64
vMotion 65
Storage vMotion 66
Physical to Virtual 67
Other VMware Datacenter Software 68
vCenter Operations Manager 68
vCenter Configuration Manager 70
Summary 70
Review Questions 72
Chapter 4 vSphere Storage 77

Physical Versus Virtual Storage 81
Direct Attached Storage 82
Network Attached Storage 83
Storage Area Network Array 83
Storage Protocols 83
Filesystem Types 84
NFS 84
vSphere Datastores 84
vSphere Datastore Clusters 85
Virtual Machine Disk Types 86
VMDK 86
RDM 87
Storage DRS and Storage I/O Control 87
SDRS 87
Storage Policies 89
Other vSphere Storage Technologies 90
vSphere Storage Appliance 90
Virtual SAN 90
vSphere Flash Read Cache 91
Virtual Volumes 91

Table of Contents xi
Availability 91
Disaster Recovery 92
vSphere Data Protection 93
vSphere Replication 94
Summary 95
Chapter 5 VMware vSphere Networking 101

Physical Versus Virtual Networking 104
VMware Virtual Switches 106
Virtual Standard Switch 106
Distributed Virtual Switches 109
Third-Party Virtual Switches 111
vSphere Networking Components 111
Portgroup 111
VMKernel Network Adapter 116
Network Input/Output Control 117
Quality of Service 119
NetFlow 119
Summary 119
Chapter 6 Virtualization Solutions That Solve Business Challenges 125

Availability Challenge Solutions 128
Solutions to Availability Challenges 128
Solving Common Business Challenges 129
Hardware Availability 129
Application Availability 129
Maintenance 130
Environmental Availability 130

xii VCA-DCV Official Cert Guide
Management Challenge Solutions 131

Solutions to Management Challenges 131
Datacenter Management 132
Provisioning Management 132
Compliance Management 133
Scalability Challenges 133
Solutions to Scalability Challenges 133
Common Scalability Challenges 134
Infrastructure Scalability 134
Virtual Machine and Application Scalability 134
Optimization Challenges 135
Solutions to Optimization Challenges 135
Monitoring Optimization 136
Efficiency Optimization 136
Performance Optimization 137
SMB Versus Enterprise Challenge Solutions 137
Size of the Company 138
Summary 139
Key Technologies 140
Review Questions 140
Answers to the “Do I Know This Already?” Quizzes and Chapter Review Questions 143
Glossary 153

About the Authors
Matthew (Matt) Vandenbeld is a dual VCDX, obtaining the certification in

both Cloud and Data center (VCDX-DCV, VCDX-Cloud). He also holds
VCAP-DCD/DCA/CID/CIA and many other industry certifications. He is cur-
rently a Solutions Architect in the Global Center of Excellence at VMware. His
role is to create solutions pertaining to the Software Defined Data Center (SDDC),
train experts in delivering SDDC solutions, and assist in high-end escalations. Matt
is very involved in the VMware community; he regularly attends VMUGs in his
region, blogs at cloudmatt.com, participates in community podcasts, and is active
on Twitter (@vcloudmatt). Matt is passionate about virtualization and assisting the
development of others in the industry.
Jonathan McDonald is currently a Senior Solutions Engineer at VMware, partici-
pating in the development of validated solution architectures, comprehensive pro-
fessional services, and scalable deep technical training as a part of the Professional
Services Engineering group of VMware Global Services. He has been with VMware
since 2005. Prior to joining Professional Services Engineering, Jonathan was a Staff
Technical Support Engineer with VMware Global Support Services, specializing in
the vCenter product and supporting technologies. In this role, he dealt with cus-
tomer escalations, mentoring, and interactions with the internal product engineering
teams. Through these different functions, he was able to provide not only reactive
support for in-depth technical problems, but also proactive support by trending
on the most commonly experienced problems and providing this information to
engineering.
Jonathan regularly speaks at VMware conferences and workshops (VMworld, Tech
Summit, VMUG, and Customer Support Days) as well as other industry events.
Jonathan is a great supporter of VMware Certifications, having completed 17
VMware technical certifications since 2006. He has also participated in the devel-
opment of VMware’s VCP- and VCAP-level content for Data center and Cloud
certifications.

Dedications
To Amanda, Victoria, and Nora. Your love and support mean the world to me and have
enabled me to succeed in my aspirations.—Matt
This book is dedicated to Audrey. May she forgive my constant working and travel in
pursuit of my career and dreams. It is not often that I look back and say, “I can’t believe I
did all of that in the last year!” This is one of those times…. Sadly, I could not find a way
to incorporate a dragon into it.—Jonathan

Acknowledgments
First, thank you to Joan Murray and Ellie Bru, who tirelessly encouraged us until
we had finished the writing and editing for this book. Joan and Ellie, thank you for
not strangling us and for providing great support to a couple of new authors who
didn’t quite realize all of the work involved in getting it out the door.
We would also like to thank our technical editor, Doug Baer. It is not an easy job
translating the technical to something that can be easily consumed. Thanks for the
insight and timely reviews.
Thank you to our managers John McGinn, Rupen Sheth, and Justin McGarry, who
have been supportive of us in this endeavor to share our knowledge.
Lastly, we would like to acknowledge all of our family and friends. It takes an
amazing amount of effort to complete a book. We thank you for the tireless hours
during which we were creating and editing. This completes a journey for us moving
an item from a “wanted to do” to a “have done” on the bucket list.
Thank you!

About the Reviewer
Doug Baer is an infrastructure architect on the Hands-on Labs team at VMware.

His nearly 20 years in IT have spanned a variety of roles, including consulting,
software development, system administration, network and storage infrastructure
solutions, training, and lab management. Doug earned a bachelor of science degree
in computer science from the University of Arizona in Tucson and holds several
top-level industry certifications, including VMware Certified Design Expert
(VCDX #19) and HP’s Master ASE Cloud and Data center Architect (#14).
You can find him working in the Hands-on Labs at VMware’s large events, pre-
senting at VMware User Group events, writing on the VMware blogs (http://blogs.
vmware.com/), or answering questions on the VMware Community forums. If you
look hard enough, you may even find him as “Trevor” in videos on the Hands-on
Labs site. In his free time, Doug likes to get away from technology and spends time
hiking with his family or running on the roads and trails all over Arizona.

We Want to Hear from You!
As the reader of this book, you are our most important critic and commentator. We
value your opinion and want to know what we’re doing right, what we could do bet-
ter, what areas you’d like to see us publish in, and any other words of wisdom you’re
willing to pass our way.
We welcome your comments. You can email or write us directly to let us know
what you did or didn’t like about this book—as well as what we can do to make our
books better.
Please note that we cannot help you with technical problems related to the topic of this book.
When you write, please be sure to include this book’s title and author as well as
your name, email address, and phone number. We will carefully review your com-
ments and share them with the authors and editors who worked on the book.
Email: VMwarePress@vmware.com
Mail: VMware Press
ATTN: Reader Feedback
800 East 96th Street
Indianapolis, IN 46240 USA
Reader Services
Visit our website and register this book at www.informit.com/title/9780133854497
for convenient access to any updates, downloads, or errata that might be available
for this book.

xviii VCA-DCV Official Cert Guide
Introduction
There is no questioning the value which virtualization has brought to the IT in-
dustry. VMware has products and technologies that revolutionize the way in which
businesses operate. From revolutionizing the way in that an administrator can
recover from failures to providing the ability to deploy a large number of servers
very quickly, the way that businesses operate is changing. The software defined
data center (SDDC) is revolutionizing the way in which IT manages the data cen-
ter. Understanding these products and technologies is instrumental to successful
implementation of them.
The value of showing your level of knowledge is second to none. Whether you
love or hate certifications, they are a necessary part of the industry. They provide
a baseline for employers to determine the level of expertise that a candidate has
in a specific area. The VMware Certified Associate—Data Center Virtualization
(VCA-DCV) certification is an entry-level certification meant to show a basic un-
derstanding of VMware data center technologies.
With the VCA-Data Center Virtualization certification, you’ll have greater cred-
ibility when discussing data center virtualization, the business challenges that
vSphere is designed to address, and how virtualizing the data center with vSphere
addresses those challenges. You’ll be able to define data center virtualization and
provide use case scenarios of how vSphere and data center virtualization can pro-
vide cost and operational benefits.
Although not a requirement to the VMware Certified Professional—Data Center
Virtualization certification, the VCA-DCV is a great starting point in your certifi-
cation journey.
Whether you simply want to become more conversant in virtualization technolo-
gies or ultimately want to be a recognized virtualization expert, VMware certifica-
tion is now an essential step for your career.

Introduction xix
Who Should Read This Book

This book is for anyone who wants to learn about VMware Data Center technolo-
gies, and is written especially for those who plan on taking the VCA-DCV exam.
We wrote this book to augment your skills and knowledge of VMware vSphere and
its associated products and to fill any knowledge gaps in preparation for the VCA-
DCV exam.
This book is aimed particularly at administrators and students who are new to
VMware virtualization technologies. It builds on prior knowledge to ensure that
you understand these technologies, and the types of business challenges which
can be addressed by them. The end goal is to prepare you to take the VCA-DCV
exam.
Goals and Methods

Our goal in writing this book is to help you prepare for the VCA-DCV Exam. It
is intended to be a basic, 100-level introduction to the concepts of vSphere and its
relating technologies. It is not intended to be an exhaustive technical or deep-dive
design guide on these subjects. As such, you may find detail lacking in certain areas
or find that you wish to learn more. In such cases, we encourage you to review the
full product documentation as well as seek out blog posts and information from the
community.
While writing this book, we adhered to the current VCA-DCV blueprint from
VMware, which can be found at the following location:
■ http://mylearn.vmware.com/mgrReg/plan.cfm?plan=41162&ui=www_cert
Please check the blueprints often, as they are updated regularly and some aspects
may change.
We did not write linearly to the blueprint. This book does however adhere closely
to the objectives in the exam blueprint. There are deviations from it and other top-
ics covered though. We tried to approach the exam topics in a way that made sense
from an administrator’s point of view. All of the exam objectives are covered in this
book and we point out the covered objectives at the beginning of each chapter. We
suggest that you read this book cover to cover as we feel that we have laid it out in
a way that makes the most sense. Once you have read the entire book and are ready
to study for the exam, read through the blueprint objective-by-objective and refer
to the corresponding sections of this book.

xx VCA-DCV Official Cert Guide
How to Use This Book

This book consists of six chapters and an appendix, detailed here. As mentioned
earlier, we have written this book from a vCloud administrator’s point of view, rather
than following the exam objectives in order.
The chapters of the book cover the following topics:
■ Chapter 1, “Datacenter Virtualization”: This chapter introduces the reader
to the concepts that are key to datacenter virtualization. This includes infor-
mation on what virtualization is, the key components in the datacenter, and
why businesses would virtualize their environment.
■ Chapter 2, “Business Challenges”: This chapter details the different types
of challenges an administrator might face in the data center. This includes
information on availability, management, scalability, optimization, and sizing
challenges.
■ Chapter 3, “vSphere Components”: This chapter details the technologies
that are important in a vSphere environment. This includes discussion on vir-
tualization, how resources are managed, high availability and disaster recovery,
migration, and other data center software from VMware.
■ Chapter 4, “vSphere Storage”: This chapter covers the key storage concepts
that are important to datacenter virtualization. It includes discussions on the
types of storage, how vSphere presents storage to a user, as well as technolo-
gies for the optimization and availability of storage in a vSphere environment.
■ Chapter 5, “VMware vSphere Networking”: This chapter provides details
on the key networking concepts that are important to datacenter virtualiza-
tion. It includes discussions on physical and virtual switches, the types of vir-
tual switches, and the networking components and technology important in a
vSphere environment.
■ Chapter 6, “Virtualization Solutions That Solve Business Challenges”:
This chapter brings everything together with a discussion comparing the com-
mon business challenges in Chapter 2, to the solutions discussed in Chapters 3
to 5. It includes a discussion on common availability, management, optimiza-
tion, and the scalability challenges that enterprises face.
■ Appendix A, “Answers to the ‘Do I Know This Already?’ and ‘Chapter
Review Questions’”: This appendix provides the answers to the “Do I Know
This Already?” quizzes that you will find at the beginning of each chapter, as
well as the “Review Questions” sections found at the end of each chapter.

Introduction xxi
Access to Practice Exams and Book Content Updates

Because VMware occasionally updates exam topics without notice, VMware Press
might post additional preparatory content on the web page associated with this book
at http://www.pearsonitcertification.com/title/9780133854497. It is a good idea to
check the website a couple of weeks before taking your exam, to review any updated
content that might be posted online. We also recommend that you periodically
check back to this page on the Pearson IT Certification website to view any errata
or supporting book files that may be available.
You will need to register this book to unlock the three practice exams that are pro-
vided with the purchase of this book.
Follow the steps below:
1. Go to pearsonitcertification.com/register and log in or create a new account.
2. Enter the ISBN: 9780133854497.
3. Answer the challenge question as proof of purchase.
4. Click on the “Access Bonus Content” link in the Registered products section of
your account page, to be taken to the three practice exams.

This chapter covers the following subjects:
■ What Is Virtualization?—This section explains what virtualization is and

how it works.
■ Datacenter Components—This section covers the components of a
physical datacenter and its virtual equivalents.
■ Why Virtualize?—This section covers the reasons virtualization is ben-
eficial to businesses.
This chapter covers a portion of the VCA-DCV objective 1.1 and objective 1.2.

CHAPTER 1
Datacenter Virtualization
The majority of server workloads now run on a virtualization platform. VM-

ware’s virtualization software is the leading virtualization platform in market
share and continues to drive innovation in the industry. In this chapter, we
discuss the basics of what virtualization is and how it offers advantages to busi-
nesses worldwide.
“Do I Know This Already?” Quiz

The “Do I Know This Already?” quiz enables you to assess whether you should
read this entire chapter or simply jump to the “Exam Preparation Tasks” sec-
tion for review. If you are in doubt, read the entire chapter. Table 1-1 outlines
the major headings in this chapter and the corresponding “Do I Know This Al-
ready?” quiz questions. You can find the answers in Appendix A, “Answers to the
‘Do I Know This Already?’ Quizzes and Chapter Review Questions.”

4 VCA-DCV Official Cert Guide
Table 1-1 “Do I Know This Already?” Foundation Topics Section-to-Question Mapping
Foundations Topics Section Questions Covered in This Section
What Is Virtualization? 1, 2
Datacenter Components 3–6
Why Virtualize? 7
1. What type of hypervisor is ESXi?

a. Type 1
b. Type 2
c. Linux-based
d. Paravirtual
2. What is a virtual machine?

a. A type of hypervisor
b. An isolated container for workloads running on a hypervisor
c. A shared operating system running on physical hardware
d. None of the above
3. Datacenters are comprised of which component(s)? (Choose all that apply.)

a. Network
b. Compute
c. Storage
d. Availability
4. What vSphere Compute feature enables vCPUs to be scheduled efficiently?

a. Math coprocessor
b. Hyperthreading
c. Relaxed co-scheduling
d. Distributed resource scheduler

Chapter 1: Datacenter Virtualization 5
5. Which two types of virtual switches are available with vSphere? (Choose two.)
a. vSphere Standard Switch
b. Consolidated Virtual Switch
c. vSphere Enhanced Switch
d. Distributed Virtual Switch
6. What types of file systems are supported as virtual machine datastores?

(Choose all that apply.)
a. VFS
b. NFS
c. CIFS
d. VMFS
e. All of the above
f. None of the above
7. What is a common business reason to virtualize physical server workloads?

a. Lower total cost of ownership (TCO)
b. Lower return on investment
c. Embracing leading-edge technologies
d. Increased performance

Foundation Topics
What Is Virtualization?
Virtualization is the abstraction of resources from its underlying hardware. The core
component of a server virtualization platform is the hypervisor. The hypervisor is a
piece of software that operates as a go-between for the guests’ operating systems
(OSes) and the bare metal hardware. Guest workloads are contained and isolated
from each other; this container is deemed a virtual machine (VM). The hypervisor
abstracts the physical hardware and separates it from the VM’s operating system.
This enables multiple VMs to operate on the same piece of physical hardware in
complete isolation. It also enables the guest to work on varying hardware platforms
without reconfiguration of the OS.
A hypervisor is installed on a physical server called a host. VMware’s ESXi hypervisor
is considered a type 1 hypervisor. These types of hypervisors do not require another
operating system to run on. Type 1 hypervisors run on bare metal hardware and
are very efficient. VMware hypervisors have become extremely efficient and highly
available. In most cases, there is no difference in performance for applications hosted
in vSphere versus running on bare metal. VMware’s most recent hypervisors run the
majority of workloads at near native performance while offering several other ad-
vantages that are covered later in this chapter.
Virtual machines, while operating on the same physical hardware, are completely
isolated from one another. Each VM is presented virtual hardware from the hypervi-
sor and operates in the exact same manner as if it were running on physical hard-
ware. Newer operating systems have embraced virtualization and are designed to
work efficiently as virtual machines. VMs send instructions to the virtual hardware
that are processed by the hypervisor and scheduled in a fair and efficient manner on
the physical hardware.
Figure 1-1 visualizes the core components of virtualization. The basic building
blocks of a datacenter are abstracted via the VMware vSphere hypervisor and pro-
vided to guest virtual machines. All components are then easily managed by vCenter
Server, which greatly increases the manageability of the entire datacenter.

VMware vCenter Server
Manage
VM VM VM VM VM VM VM VM VM VM VM VM
VMware vSphere VMware vSphere VMware vSphere
Figure 1-1 The abstraction performed by the hypervisor
Datacenter Components
In this section, we discuss the core components of the virtual datacenter. These
components may differ slightly from other datacenter definitions.
The datacenter components that form the components of a virtual datacenter are as
follows:
■ Compute
■ Network
■ Storage
Server virtualization, via the hypervisor, abstracts these elements and presents virtual
equivalents to the virtual machine’s operating system. Without these core compo-
nents, a VM will not function. Although power, cooling, racking, cabling, and other

components might be referenced, they are outside the core virtual datacenter defini-
tion.
Compute
Datacenter compute components are comprised of central processing unit (CPU)
and memory resources. These resources facilitate the computational requirements
for applications. CPU and memory are housed within a physical server. This physi-
cal server becomes the host for the ESXi hypervisor. ESXi takes control of all of
the scheduling for CPU and memory resources. It efficiently and fairly shares these
resources amongst many virtual machines.
An instruction sent from a virtual machine to its virtual CPU (vCPU) is processed
by the hypervisor and scheduled on a physical CPU core. Multiple cores and mul-
tiple vCPUs can be scheduled to run concurrently, or rather, at the same time. This
concept is referred to as co-scheduling. vSphere utilizes a type of scheduling known
as relaxed co-scheduling. This type of co-scheduling flexibly schedules the vCPUs of
a guest in an intelligent manner that is an improvement in the hypervisor schedul-
ing algorithm of other hypervisors. This greatly increases the efficiency of multiple
workloads on a single piece of hardware. The hypervisor uses the concept of shares
to prioritize access to CPU and memory resources. These shares are customizable
and are used in the resource entitlement calculation. When a host is not busy and
there are CPU cycles to spare, each VM’s entitlement will equal its demand. If, for
example, all CPU cores within a host are busy because many busy VMs are trying
to use the same physical CPUs, the hypervisor uses the entitlement calculation to
determine which workload to run next or whether to interrupt a running process for
another.
Memory is allocated within the hypervisor to each VM to maintain isolation; this
allocation uses a hypervisor data structure called a pmap, which maps the “physical”
memory that a VM sees to actual physical host memory. This mapping enables the
guest OS to see a contiguous block of addressable memory, even though the actual
memory blocks may be distributed across the host’s physical memory. The ESXi hy-
pervisor uses an entity called the shadow page table to maintain virtual-to-physical
memory mapping. ESXi scans memory to find duplicate entries; these entries are
then consolidated into a single entry and pointers are created in place of the du-
plicates. This allows for oversubscription of memory resources and more efficient
use of memory as a whole. This is one of the techniques ESXi utilizes to manage
memory, called transparent page sharing. During times of contention, ESXi can use
memory reclamation mechanisms such as ballooning, swapping, and compression
NOTE Compute is covered in depth in Chapter 3, “vSphere Components.”

Network
Datacenter networking facilitates the communication between devices both inside
and outside the datacenter. Packets are transported from the source to the destina-
tion. Dependent on where both the source and destination are located, the packet
will be moved through varying devices, both physical and virtual. The network may
consist of routers, switches, and adapters at the physical layer. Networks can also be
segmented at a logical layer into different network segments. Virtual local area net-
works (VLANs) are an example of this.
At the virtual layer the hypervisor creates virtual switches to enable VMs to commu-
nicate with the network. There are two different types of virtual switches—vSphere
standard switch (vSS) and virtual distributed switch (vDS). vDS can also be referred
to as distributed virtual switches (DVS).
Virtual network adapters are presented by the hypervisor to the guest operating
system. There are multiple types of adapters to ensure performance and support for
multiple operating system types. The guest utilizes this virtual network adapter to
send and receive packets via the hypervisor. The hypervisor controls all packets from
its physical adapters through the virtual switch to the guest, and vice versa.
NOTE Network is covered in depth in Chapter 5, “VMware vSphere Networking.”
Storage
Storage devices contain all data within a datacenter, including operating systems
and VMs. Data integrity and reliability are paramount within datacenters because
storage corruption or unavailability will cause major service outage. Although data
integrity is paramount, performance must not be overlooked. Storage is the most
common cause of performance-related issues within a datacenter. A balance between
data protection and performance must be achieved. The storage technology used to
achieve these goals is a Redundant Array of Independent Disks (RAID). RAID tech-
nology uses many hard disk drives to offer protection from individual disk failure
as well as offer performance via multiple devices operating concurrently. There are
multiple RAID levels, with the most common being these:
■ RAID0 Striping—Striping between drives for performance; no redundancy
for disk failures. Requires at least two drives.
■ RAID1 Mirroring—Mirroring between drives; slower performance but will
tolerate drive failures. Requires at least two drives, and drives must be added in
pairs.

■ RAID5 Parity—Requires a minimum of three drives and is a common balance

between performance and redundancy with the use of distributed parity. This
allows for a single drive failure while maintaining data access via the operating
drives.
Multiple other RAID types are available depending on the type of storage device
you are using. Familiarize yourself with your particular vendor’s RAID offerings to
ensure you have selected the appropriate level for your requirements.
RAID arrays can be located locally within servers or on shared storage devices. Stor-
age devices within the datacenter vary depending on the size of the organization and
the associated budget. The most common types are direct attached storage (DAS),
network attached storage (NAS), and storage area network (SAN). The most com-
mon example of direct attached storage is the disk drives located within servers.
NAS and SAN devices are shared storage devices that can be accessed by multiple
systems concurrently. NAS and SAN devices are very similar in nature—in some
cases interchangeable. The primary differentiator is how the storage is made avail-
able to hosts. SAN presents block-based storage, whereas NAS presents file-based.
These storage devices store data in one of two ways, at the file level or at the block
level. File-level storage requires the storage array to provide hosts with a file system
for them to utilize; the most common of these are NFS and CIFS. Block-level stor-
age provides hosts with raw storage—storage without a file system—the hosts them-
selves then write their own file system. ESXi formats raw devices with the VMFS
file system.
vSphere supports two types of file systems: NFS and VMFS. There are advantages
and disadvantages to both. Neither system is better than the other; the selection of
the file system is heavily dependent on your design requirements and constraints.
vSphere hosts then use these file systems as datastores. Datastores are the location
the hosts use to store VM data. vSphere datastores can be shared amongst many
hosts. This enables many of the vSphere availability features and is a key component
of the virtual datacenter.
NOTE Storage is covered in depth in Chapter 4, “vSphere Storage.”

Why Virtualize?
Virtualization has revolutionized the IT industry. In a very short time, it has gone
from small dev/test environments to enterprise tier 1 datacenters. The majority of
workloads worldwide now run on a virtualized platform, and this ratio will only in-
crease in the future. Virtualization is also the foundation for cloud platforms, which
is the future of the IT landscape. What is it about virtualization that has caused such
a rapid transformation of the IT industry? There are an incredible number of rea-
sons to virtualize; included here are a few of the most prominent reasons.
Optimization
The initial motivator to move to virtualization was server consolidation. Server
sprawl was a very big problem in the “1 app, 1 server” datacenter environment.
Hundreds of servers were using power, generating heat, taking up space, and requir-
ing administration. On top of that, the great majority of these servers were woefully
underutilized, using just a fraction of their CPU and memory. It was a completely
inefficient way to run a datacenter. Virtualization allowed one physical server to be
shared by multiple VMs. In the early days, you could get ratios of 20 to 1—this rep-
resented significant cost savings to the business. Over time, these ratios have grown,
and it is not uncommon to see ratios of 100 to 1 now.
These consolidation ratios provide a high return on investment (ROI) for purchased
physical equipment and lower the total cost of ownership (TCO) for an organiza-
tion. This is one of the primary motivators for businesses to embrace virtualization.
Virtualizing physical servers saves money that would have been spent on more phys-
ical servers than required, and gets full utilization from servers already purchased.
There are some situations in which high consolidation ratios are not preferred. This
is the case in tier 1 or mission-critical applications. These applications are being vir-
tualized to take advantage of the availability provided within a vSphere environment.
Figure 1-2 illustrates the optimization and consolidation of datacenter resources.

APP
OS
APP
APP
OS
OS APP
APP
OS
OS
APP
OS
Figure 1-2 An entire rack of servers consolidated onto a single vSphere server
Availability
vSphere enables applications to be migrated between hosts without interruption to
service, which is highly advantageous and allows for server maintenance in a greatly
simplified manner. The latest iteration of vMotion (dubbed Enhanced vMotion)
enables live migration without the requirement of shared storage. This allows you
to move workloads away from essentially any device to perform maintenance or up-
grades without impact.
In the event of unplanned outages, vSphere uses the High Availability (HA) feature
to automatically restart workloads on functioning hosts in the cluster. This greatly
reduces the impact of host failures within a datacenter. In traditional datacenters,
a host failure could interrupt application availability for hours to days depending
on the failure. However, with vSphere HA, this outage lasts a matter of minutes.
HA is also configurable, and you can select certain VMs to have a higher or lower
restart priority depending on your requirements. The newest iteration has added

application-aware HA within vSphere for select applications. This enables vSphere

to restart, for example, the SQL service on a database server if it fails.
If a few minutes of outage is too much to tolerate, vSphere offers fault tolerant
(FT) VMs. Fault tolerance enables an application to stay online even through a host
failure event. This is accomplished by mirroring a VM on two separate hosts at the
same time. It uses vLockstep technology to replicate all changes to the replicated VM.
At this time, there are fairly tight restrictions on the VM configuration available for
FT.
Figure 1-3 shows VMware vSphere high availability.
VM VM VM VM VM VM
Resource Pool
VMware ESXi VMware ESXi VMware ESXi
Operating Server Failed Server Operating Server
Figure 1-3 Workloads moving automatically to available servers from a failed server
Management
Managing hundreds of physical servers is an onerous task further complicated by
differing and expensive management tools. VMware’s vCenter Server greatly reduces
administrative effort to manage large environments. In traditional datacenters, most
organizations tried to maintain a ratio of 1 administrator to every 25 servers. Today,
a single administrator can look after hundreds of virtualized servers with ease. This
frees up resources to drive even more projects and innovation within an organiza-
tion as opposed to just keeping the lights on.
To further simplify operations management, vCenter will automatically relocate
resources in times of contention to ensure the workload is evenly distributed. This
function is handled by the Distributed Resource Scheduler (DRS) and is highly con-
figurable and robust. The DRS concept has been extended to the storage subsystem
as well and can manage the utilization of datastores in the environment. Storage
DRS (SDRS) ensures VMs are placed onto datastores with enough capacity and

periodically checks the latency of the datastore. If specified thresholds are exceeded,
vCenter moves the VM to another datastore that will meet its requirements.
Figure 1-4 shows vMotion that greatly eases management tasks.
VM
wa
re
ES
Xi
vM
oti
on
VM
wa
re
ES
Xi
Figure 1-4 vMotion live migrating a running VM from one host to another
Scalability
Provisioning times for servers have been drastically reduced from weeks to minutes.
To provision a physical server, it must be procured, shipped, received, unboxed,
racked, cabled, an OS deployed to it, custom drivers installed, and security patched.
To provision a virtual server, an administrator can right-click a template and select
Deploy VM from This Template. It’s really that simple—as long as templates have
been created and are up-to-date. This incredible ease of provisioning and the stan-
dardization of templates have enabled the cloud principle of self-provisioning.
To scale up physical servers in the traditional datacenter, there must be an abun-
dance of room (rack space), cooling, and power. In addition, complexities exist
around procuring physical servers and their build times. With vSphere, scaling is
core to the product. Virtual machines can be rapidly provisioned and managed easily.
A greatly reduced number of physical servers are required, and ESXi can be rapidly
deployed to new hardware, especially with the use of Auto Deploy and host profiles.
These tools enable you to provision new hosts within minutes of hardware arriving
onsite; these hosts then can host many VMs.

Figure 1-5 demonstrates the scalability of the vSphere platform.
VM
oti
on
VM
wa
re
VM
wa
re
VM
wa
re
Figure 1-5 Hosts can be easily added to accommodate VM growth
NOTE Business challenges are also covered in Chapter 2, “Business Challenges.”
Summary
In this chapter, you have learned the basics of what virtualization is and what its key
components are. We explained the role of the hypervisor in abstracting physical
hardware. The concept of VMs and their functionality was introduced. These are
the core concepts that build to an understanding of what virtualization is and how it
works.

The components of the datacenter were introduced in this chapter. Compute, net-
work, and storage are the building blocks utilized by virtualization. Each component
is abstracted by the hypervisor and made available to VMs to use.
The reasons virtualization are used by business were also explored. The key con-
cepts of optimization, availability, management, and scalability were explained. VM-
ware virtualization offers many benefits to businesses and has become an integral
part of the majority of datacenters.
Exam Preparation Tasks
Review All the Key Topics

Table 1-2 provides a detailed discussion of the key topics. Use this table as a quick
reference to topics required by the blueprint. Examples of these and other settings
are provided in the following sections.
Table 1-2 Key Topics for Chapter 1
Key Topic Description Page Number

Definition Definition of a hypervisor. 6
Definition Definition of a virtual machine (VM). 6
List Datacenter components that form the components of a 7
virtual datacenter.
Definition Definition of a vCPU. 8
Definition The two types of virtual switch are defined. 9
Definition The two types of supported file systems are defined. 10
Definition Return on investment (ROI) and total cost of ownership 11
(TCO) are defined.
Figure 1-3 HA recovering from a failed server. 13
Figure 1-4 vMotion migrating VMs from one host to another. 14

Definitions of Key Terms

Define the following key terms from this chapter, and check your answers in the
glossary.
Hypervisor, Compute, Network, Storage, Optimization, Availability, Management,
Scalability
Chapter Review Questions

These review questions validate that you understand the topics covered in this chapter.
1. Which of the following is a type 1 hypervisor?
a. Workstation
b. ESXi
c. Fusion
d. vCloud Director
2. What is an isolated container running an application on a hypervisor?

a. A virtual container
b. A virtual machine
c. A datastore
d. None of the above
3. What is a component of a datacenter?

a. Network
b. CPU
c. Quality of service
d. Availability
4. What is the vSphere memory feature that consolidates duplicate memory

pages?
a. Transparent page sharing (TPS)
b. Hyperthreading
c. Relaxed co-scheduling
d. Distributed resource scheduler

5. Which type of virtual switch includes advanced functionality available with

vSphere?
a. Virtual Standard Switch
b. Consolidated Virtual Switch
c. vSphere Switch
d. Distributed Virtual Switch
6. Which file system does vSphere deploy on block-based storage?

a. VFS
b. NFS
c. CIFS
d. VMFS
7. What is a common business reason to virtualize?

a. Higher total cost of ownership (TCO)
b. Higher return on investment (ROI)
c. Embracing leading-edge technologies
d. Increased performance
8. Which vSphere feature greatly reduces VM deployment time?

a. Templates
b. DRS
c. Hyperthreading
d. Snapshots
9. Which vSphere feature enables you to migrate a VM to a new host and data-
store at the same time?
a. vMotion
b. Storage vMotion
c. Multi vMotion
d. Enhanced vMotion

10. Which vSphere feature enables the rapid provisioning of vSphere hosts?
a. Auto Deploy
b. DRS
c. Templates
d. Clones

■ Availability Challenges—This section discusses challenges businesses

have in ensuring that resources are available.
■ Management Challenges—This section discusses the challenges busi-
nesses have in ensuring resources can be effectively managed.
■ Scalability Challenges—This section discusses the challenges businesses
have in ensuring that the datacenter can scale to meet the demand.
■ Optimization Challenges—This section discusses the challenges busi-
nesses have in ensuring that they are using the environmental resources
effectively.
■ SMB Versus Enterprise Challenges—In this section, you learn about
various challenges faced by small and medium businesses (SMBs) and en-
terprises that are dependent on the size of the environment.
This chapter covers VCA-DCV objective 1.2.

CHAPTER 2
Business Challenges
When you are operating a business, a primary concern is the common busi-
ness challenges that need to be taken into account to ensure that the day-to-
day operations are smooth and uninterrupted. In this chapter, you learn about
the types of challenges businesses face while running a datacenter. Availability,
management, scalability, and optimization of the datacenter resources are of
utmost importance to ensuring that the cost of doing business remains as low as
possible. Understanding the challenges also ensures that the recovery time and
recovery point objectives (RTO and RPOs) can be flexible and met with ease.
These challenges can vary from small businesses to large enterprises; however,
all businesses are looking to achieve the same basic needs, only on a different
scale. Specific technologies offered by VMware that enable you to meet these
business challenges are not discussed until Chapter 6, “Virtualization Solutions
That Solve Business Challenges.”



Availability Challenges 1, 2
Management Challenges 3, 4
Scalability Challenges 5, 6
Optimization Challenges 7, 8
SMB Versus Enterprise Challenges 9, 10
1. Which of the following is an availability challenge?

a. Centralized management of the virtualized environment
b. Needing detailed performance reports to trend on resource utilization
c. Performing maintenance during normal business hours
d. Not having storage in the environment
2. Wanting to know whether you have performance bottlenecks is an availability

challenge.
a. True
b. False
3. Which of the following is a management challenge?

a. Prioritization of network bandwidth for specific traffic
b. Needing a solution to monitor for regulatory compliance
c. Needing an automated system to correct unbalanced resources during
growth periods
d. Needing the ability to quickly test a disaster recovery plan
4. Wanting to reduce the power and cooling requirements of the datacenter is a

management challenge.
a. True
b. False

Chapter 2: Business Challenges 23
5. Having the ability to add more RAM is a scalability characteristic.

a. True
b. False
6. Which of the following is a scalability challenge?

a. Needing the ability to quickly expand the environment in a short time
b. Prioritization of network bandwidth
c. Minimizing the downtime caused by hardware failures
d. Wanting to use disk space more efficiently
7. Which of the following is an optimization challenge?

a. Needing the ability to quickly test a disaster recovery plan
b. Performing maintenance during regular business hours
c. Wanting centralized management of resources.
d. Needing detailed performance reports to trend on resource utilization
8. Wanting to satisfy an aggressive service level agreement is an optimization

challenge.
a. True
b. False
9. Which characteristics will have a larger impact for enterprises than SMBs?
(Choose all that apply.)
a. Number of failover hosts available
b. Ensuring management network availability for hosts
c. Scale of the infrastructure
d. Number of virtual machines
10. Enterprises and SMBs are managed in a similar way.

a. True
b. False

Foundation Topics
Availability Challenges
To businesses, being highly available is of utmost concern. In many cases, even a lit-
tle bit of downtime can cost millions of dollars. This section discusses what availabil-
ity is, as well as the various types of challenges customers face in the environment.
What Is Availability?
Availability in its simplest form is the amount of time that a resource or service is
available to be consumed. This could be in terms of an application, an entire server,
or even an entire site. It is all relevant to ensuring that users are able to access the
resources they need to do their jobs. Figure 2-1 shows an example of a process flow
for ensuring an application is available.
2 APP X1 APP
3 OS OS
Managed Hosts
Figure 2-1 Process for application availability
In this case, the application has a failure (1), thus a restart on the application is at-
tempted first (2), and then the operating system is completely rebooted in an at-
tempt to restore application availability (3). This demonstrates a workflow that could
be desired if there is a failure. Failures often are unique and need to be addressed in
completely different ways, depending on the cause of the issue.
Availability is also very important to the recovery time objective (RTO) and recovery
point objectives (RPOs) of the organization. Thus, an appropriate solution has to be
in place to cover many types of failures.
When looking at different types of availability, you can break it down into these cat-
egories:
■ Hardware—Where an administrator needs to take into account the potential
for failure of physical hardware and how the environment reacts to a server
that is physically having a component failure. Generally, these errors are cata-
strophic and need to be addressed before operation can continue normally.

■ Application—Where the application needs to have recoverable features for

detecting a failure. Generally, these errors are fixed by a service or server re-
start (whether automatic or manual) or by steps inside the code of the applica-
tion.
■ Maintenance—Where the administrator needs to be able to perform main-
tenance but maintain the availability of resources in the physical environment
with as little downtime as possible.
■ Environmental—Where the administrator needs to ensure that environmen-
tal issues (disaster, power failure, and so on) enable the business to continue
functioning, even if it is at a reduced capacity. Generally, this results in recov-
ery or failover to a different site or datacenter that is not affected by the issue.
Common Business Challenges

Given the different types of availability, there are common business challenges that
map to them. In this section you learn how to identify challenges for the four cat-
egories of availability:
■ Hardware
■ Application
■ Maintenance
■ Environmental
Hardware Availability
Common hardware availability challenges often sound like the following statements:
■ “My company needs to minimize the downtime caused by a host failing in the
environment.”
■ “My company has a service level agreement that calls for a very quick response
to a host failing. Therefore, I have to be sure that virtual machines are up and
running as quickly as possible after a failure.”
■ “I need to ensure that if a host loses network access temporarily, appropriate
action is taken.”

Application Availability
Common application availability challenges often sound like the following state-
ments:
■ “We have an application that cannot be interrupted by hardware failures.”
■ “I need to ensure that if my application stops responding, action is automati-
cally taken to try to resolve the problem.”
■ “My company requires that if an operating system stops responding for a pe-
riod of time, action is automatically taken to try to resolve the problem.”
Maintenance
Common maintenance challenges often sound like the following statements:
■ “I need to be able to perform hardware maintenance at any time without im-
pacting the availability of services.”
■ “We need to avoid taking multiple servers down to apply hotfixes.”
■ “My company needs to perform hardware updates at any time of the day with-
out impacting users who are working.”
Environmental Availability
Common environmental availability challenges often sound like the following state-
ments:
■ “Our process to back up servers is difficult. It is hard to test for validity to
ensure that they are recoverable, and it is hard to restore an entire server
quickly.”
■ “My company has mission-critical applications that must be available even if
environmental conditions cause problems at the datacenter. We would like a
solution for this that will also allow us to easily test the environment to be sure
that any action that is taken will be successful if we do have a disaster.”
■ “We would like to replicate server data to another site but have different types
of storage array at each site.”

Management Challenges
Being able to manage a large environment is almost always a challenge for adminis-
trators. In this section, you learn about management and the challenges administra-
tors often face in managing an environment.
What Is Management?
Management is the ability to effectively administrate the environment and easily
work with any system in it. In addition, a single pane of glass into the environment
is a goal many companies have to address the complexities of managing an environ-
ment. Management includes such things as being able to quickly provision new sys-
tems; managing an overcrowded datacenter, roles, and responsibilities; and tracking
and monitoring for changes to the environment. This is shown in Figure 2-2.
Management Server
Manage
Host Host Host
Figure 2-2 Environmental management
Centralized management as shown is critical to effectively and efficiently managing

an environment. Having the ability to effectively manage an environment is impor-
tant to providing a flexible and compliant environment that is capable of meeting
the requirements of the organization.
When looking at datacenter management, it can be broken down into the following
categories:

■ Datacenter—Focuses on consolidation and centralized management of the

datacenter. In many cases, the datacenter is running out of space, or it is dif-
ficult for an administrator to interact with the environment.
■ Provisioning—Focuses on creation or migration of resources in the environ-
ment. Often in traditional datacenters, quickly provisioning resources is dif-
ficult due to the complexity of change control impacting the environment.
■ Compliance—Focuses on adherence of the environment to a standard.
Commonly businesses need to comply with regulatory standards such as the
Payment Card Industry Data Security Standard (PCI DSS) or Healthcare
Insurance Portability and Accountability Act (HIPAA), which can create addi-
tional complexity in the environment.

Given the various types of management challenges, there are common business chal-
lenges that map to them. In this section, you learn how to identify challenges for the
three categories of management discussed. These include
■ Datacenter
■ Provisioning
■ Compliance
Datacenter Management
Common datacenter management challenges often sound like the following state-
ments:
■ “I am the only administrator for the environment, and I need a centralized way
to manage my datacenter.”
■ “My datacenter is almost out of space. Power and cooling are maxed out, and it
is difficult to introduce new hardware into the environment.”
■ “Our environment is complex, and we have many administrators around the
world. We need to be able to see everything around the world in a single pane
of glass for administrators.”

Provisioning Management
Common provisioning management challenges often sound like the following state-
ments:
■ “We have many physical servers running on really old hardware. This hard-
ware is out of warranty, and the applications are difficult to reinstall.”
■ “I need to be able to provision a server from a standardized image that my or-
ganization uses.”
■ “My company is growing quickly, and we need a way to rapidly deploy many
virtual machines to meet the demand.”
Compliance Management
Common compliance management challenges often sound like the following state-
ments:
■ “My company is in retail, and we have a need to ensure we meet compliance
regulations for credit card information.”
■ “We need to monitor our environment to make certain that our servers are
running the latest patches. If they are not, an administrator needs to be noti-
fied.”
■ “I need to monitor an environment for unexpected changes. This will help to
determine whether change control has been bypassed for critical servers.”
Scalability Challenges
Scalability in an environment will in most cases dictate how much can be done with
a particular set of hardware. In this section, you learn about scalability and the chal-
lenges administrators face.
What Is Scalability?
Scalability is the ability of the infrastructure to handle the ever-growing load that is
generated as a result of doing business. This load could be as a result of increased
demand for resources (tax season or Black Friday are examples) or with the through-
put of an application. Scalability is often brought up in capacity discussions because
it is pertinent in discussions about load and increasing demand for resources in the
environment. Figure 2-3 shows an environment that has scaled to meet the demand
on it.

Hosts
APP
OS
APP
APP
OS
OS APP
APP
OS
OS
APP
OS
Figure 2-3 System scalability
Scalability is a constraint of any environment. It does not take a long time before a
single server turns into multiple racks of servers or a single application needs to be
scaled out in an attempt to increase the scalability required for the environment. Be-
ing able to determine and assess the environment for these conditions is incredibly
important.
Scalability is also important to the RTOs and RPOs of the organization. If you are
unable to meet the objectives due to resource constraints, the consequences can be
catastrophic to the business. Therefore, the systems need to scale to meet the de-
mand of the business but still need to be able to handle an outage or an expanded
load. Many times the disaster recovery plan is last to be thought of until it is too late.
When looking at scalability, it can be broken down into these categories:
■ Infrastructure scalability—The ability of the underlying infrastructure to
handle the needs of the servers in the infrastructure. Generally, this is things

such as the amount of shared storage available, whether to use thin disk provi-
sioning, or the throughput on the network ports available in the environment.
■ Virtual machine and application scalability—The ability to handle the in-
creased demand for compute resources in the environment. Generally, this in-
volves the ability to add resources on demand, to scale out existing applications
easily, and to track where resources are being used.

Given the various types of scalability, there are common business challenges that
map to them. In this section you learn how to identify challenges for the categories
of scalability:
■ Infrastructure
■ Virtual machine and application
Infrastructure Scalability
Common infrastructure scalability challenges often sound like the following state-
ments:
■ “We are a small company and do not have a SAN. We still want to be able to
take advantage of virtualization.”
■ “My company has the need to scale out the network bandwidth available to my
virtual hosts without taking downtime on the virtual machines.”
■ “I need to ensure that I can control the characteristics of the storage that is
used to ensure that I have enough room to scale my environment appropri-
ately to meet the needs of my servers.”
Virtual Machine and Application Scalability

Common server scalability challenges often sound like the following statements:
■ “My company has an application that requires dedicated servers and will be ex-
panding due to a merger with another company this year. We need to be able
to scale out this application quickly.”
■ “We need the ability to add more CPU or RAM to the servers without having
to take the servers down.”
■ “We need an automated system that will detect and correct issues with re-
source utilization on our hosts or storage arrays.”

Optimization Challenges
Ensuring that the environment is able to effectively operate given a set of parame-
ters is key to ensuring that resources are not being ineffectively used. In this section,
you learn about environmental optimization and the challenges administrators face
in regards to it.
What Is Optimization?
Optimization is the ability to recognize inefficiencies in an environment and take ac-
tion to correct them. This enables a workload to take full advantage of the available
performance of the environment. Optimization can take place either automatically
or manually. Figure 2-4 shows a virtual machine (VM) migration taking place to bet-
ter optimize the environment by balancing the number of VMs across all hosts in
the cluster.
VM VM VM VM VM VM VM VM VM
Figure 2-4 Optimizing resources
In Figure 2-4, the system detects that one host is underused as compared to the
other hosts in the cluster. A migration (1) is done to distribute the load more effec-
tively to optimize the performance of the overall system.
Automated detection and remediation of the environment enable the environment
to be proactively monitored and action taken where necessary. This could be as sim-
ple as alerting the administrator or more complex, such as performing a migration.
Resource utilization is a common area where administrators often can benefit from
proactively monitoring the environment in this way.

When reviewing optimization challenges, they can be broken down into these cat-
egories:
■ Monitoring—Active review of the environment to determine the state of the
environment. This includes areas such as health, efficiency, and risk. These are
key areas that are important to monitor for negative future impact.
■ Efficiency—The ability to more effectively use resources to meet the growing
demand of the business without negatively impacting other servers.
■ Performance—Ensuring that the environment is optimized for peak perfor-
mance as needed by the environment. This includes prioritization of specific
resources as required by the business.

Given the different types of optimization, there are common business challenges
that map to them. In this section, you learn how to identify challenges for the three
categories of optimization:
■ Monitoring
■ Efficiency
■ Performance
Monitoring Optimization
Common optimization monitoring challenges often sound like the following state-
ments:
■ “We would like to be able to proactively monitor for performance bottlenecks
in our environment. Currently our environmental monitoring provides only a
few basic statistics that we can look at.”
■ “My company needs to have detailed performance reports that can be used to
trend our environment.”
■ “I have several servers that are idle most of the time. I need a way to monitor
them and assess whether consolidating them will affect performance.”
Efficiency Optimization
Common efficiency optimization challenges often sound like the following state-
ments:

■ “We have several servers that have very little disk usage but were configured
with a large amount of disk space. It would be great to be able to minimize the
impact this has on the overall disk usage.”
■ “I need a way to detect whether I am efficiently using the available resources
in my environment.”
■ “My company has several different storage arrays that are used in the environ-
ment. Some of these storage arrays are high performance and cost a lot more
to operate than others. Many times development users place virtual machines
on the incorrect storage device, causing degradation of performance for the ar-
ray, which we correct by moving the virtual machine to the right class of stor-
age. We need an automatic way to ensure that users can choose the appropriate
class of storage from the time of deployment.”
Performance Optimization
Common performance optimization challenges often sound like the following state-
ments:
■ “I need a way to proactively ensure that I have a healthy environment so that I
can get optimal performance of my hosts.”
■ “We need a way to prioritize network traffic to ensure that our mission-critical
virtual machines get the peak performance possible.”
■ “My company needs to ensure that we can add resources to virtual machines in
our environment on-the-fly to address performance problems if they occur.”
SMB Versus Enterprise Challenges

The size of the company can have an impact on the challenges that might be faced
in the environment. In this section, you learn about how the size of the business af-
fects the decisions that must be made to ensure the environment is healthy.
Size of the Company

The challenges faced by small and medium-sized businesses and large enterprises
are very similar. However, the solutions often differ dramatically. All companies are
interested in the availability, scalability, optimization, and manageability of their en-
vironment. The size of a company, though, often makes the choice of a solution very
different. This could be due to the available hardware in the environment or to the
number of servers that need to be backed up.

Most solutions can, with some predictability, be used in any environment; others
are not suited for certain environments or for certain levels of service. For example,
vSphere Replication is a suitable technology to implement only if the RPO of the
business is less than 15 minutes. Knowing when a solution fits a particular business
size or use case is incredibly important to ensuring a properly operating environ-
ment.
In addition, the total amount of data that must be managed often greatly differs
given the size of the environment. If the rate of change of data is greater than the
possible bandwidth available to replicate that data, then it is not going to be able
to meet the needs of the company. Thus, another solution will be needed. Whether
that solution is to add more bandwidth or to be more selective as to which data is
replicated, there must be a compromise. This compromise is typically in the form of
cost versus function because the price for some solutions might not be viable for all
companies.
The following is a list of common challenges that might need to be taken into ac-
count when assessing the environment:
■ Amount of available bandwidth—In smaller businesses, the bandwidth con-
cerns might not be as much of an issue as in a larger enterprise where there is
a heavier workload. The cost of additional bandwidth can be more than is pos-
sible for the function being performed.
■ Backup strategy—It is easy to build a solution that says “Back up the entire
environment every night”; however, this might not be possible in an enterprise
that has several thousand VMs. Often the amount of time becomes prohibitive
to this type of solution.
■ Availability strategy—In an enterprise, it is a lot easier to ensure that there
are available resources to recover from a failure. Whether the failure is a single
host or an entire site, resources are available. In a SMB, however, losing a
single server could be reducing capacity by a significant amount. Thus, appro-
priate solutions and policies must be set to ensure proper recovery.
■ Centralized management—It is easy to have everything centrally managed in
a smaller business because fewer hosts typically must be managed. When you
are an enterprise business with multiple sites and many more servers, this is
a lot more difficult to accomplish. In multisite scenarios, network bandwidth,
latency, and redundancy become larger concerns than in smaller, single-site
environments.
■ Compliance and configuration management—In an SMB, keeping tabs on
environment changes that would violate compliance regulations can be easier,
and specialized software might not be required. In enterprise environments,

software such as vSphere configuration manager (VCM) are a must because

with thousands of hosts and virtual machines, it is next to impossible to quickly
validate compliance.
These are only a few examples to make you start to think about the types of differ-
ences between SMBs and enterprise environments. Some might amount to nothing,
while others might end up being a deal-breaker for the technology.
Summary
In this chapter, we discussed the various types of challenges an administrator must
consider when designing and implementing an environment. Availability, manage-
ment, scale, and optimization of the environment are key to being able to design an
infrastructure that meets the standards for the business:
■ Availability enables an administrator to take into account any environmental
failures, from a single server up to the entire site. Ensuring that you are able to
meet the RPO or RTO appropriately for the workload will dictate how avail-
able the environment needs to be.
■ Management is specific to managing the resources in the environment. Cen-
tralizing the management of resources and the ability to perform maintenance
without impact to resources are often the cardinal needs of the business.
■ Scalability of the environment is all about what is needed to be able to do the
work in the environment. In many cases, this equates to how much work can
be done in the datacenter before resources are taxed to the maximum possible.
■ Finally, optimization is the ability to monitor the environment for the health
of the environment and any anomalies that might occur during regular opera-
tions. The datacenter will have peaks and valleys in terms of utilization, to
which an administrator needs to ensure that the health does not decline.
How these challenges can be met is discussed in Chapter 6, after you learn how vir-
tualization concepts can be applied to meet these challenges


reference to the settings you need to make or verify in any system. Examples of
these and other settings are provided in the following sections.
Key Topic Element Description Page

List The types and challenges associated with availability 25
List The types and challenges associated with management 28
List The types and challenges associated with scalability 31
List The types and challenges associated with optimization 33
List The types and challenges associated with smaller versus bigger 35
businesses

glossary.
Availability Challenges, Management Challenges, Scalability Challenges, Optimiza-
tion Challenges, SMB Versus Enterprise Challenges
Review Questions
The review questions section is for you to review your knowledge of the topics in
the chapter. You can find the answers in Appendix A.
1. Which of the following is an availability challenge?
a. Centralized management of a few hosts.
b. Ensuring that downtime is minimized.
c. We want to use our disk space as efficiently as possible.
d. Trending on resource utilization.

2. Your manager has asked you to ensure that the mission-critical virtual ma-
chines in the environment are protected in case of environmental disaster.
Which type of challenge is this?
a. Scalability
b. Management
c. Optimization
d. Availability
3. Which of the following is a management challenge?

a. IT operations require a single pane of glass view for the servers available
in the environment.
b. Monitoring the environment for bottlenecks.
c. Needing an automated system to correct unbalanced resources during
growth periods.
d. Needing the ability to quickly test a disaster recovery plan.
4. Your boss wants to reduce the power and cooling requirements of the datacen-
ter. Which type of challenge is this? (Choose all that apply.)
a. Management
b. Scalability
c. Optimization
d. Availability
5. You need to be able to add more RAM to a VM. Which type of challenge is
this?
a. Optimization
b. Availability
c. Scalability
d. Management
6. Which of the following is a scalability challenge?

a. Prioritization of storage bandwidth
b. Centralized management of resources
c. Minimizing the downtime caused by hardware failures
d. Small sites that do not have any available space to add additional CPUs

7. Which of the following is an optimization challenge?

a. Viewing how efficiently the datacenter is operating
b. Growing the environment very quickly
c. Hardware maintenance for a failing server
d. Having a single pane of glass view for the environment
8. You need to ensure that the datacenter is healthy. Which type of challenge is
this?
a. Scalability
b. Optimization
c. Management
d. Availability

■ Virtualization Concepts—This section explains the basic tenets of virtu-
alization, including virtual resources, virtual machines, and hypervisors.
■ Managing Resources—In this section, you learn about how resources are
managed in a VMware vSphere environment.
■ High Availability and Disaster Recovery—In this section, you learn
about how environments can be made highly available and protected in
the case of a failure.
■ Migration—In this section, you learn about the migration of virtual re-
sources.
■ Other VMware Datacenter Software—In this section, you learn about
other datacenter software that helps to monitor and maintain an environ-
ment.
This chapter covers VCA-DCV objective 2.1 and a portion of VCA-DCV

objective 1.1.

CHAPTER 3
vSphere Components
The foundation of any vSphere environment is the components that comprise

the virtual datacenter. In this chapter we discuss the core virtualization concepts
and constructs. This includes virtual machines (VMs) and hypervisors, as well
as the technologies involved in managing resources, high availability and disas-
ter recovery, and migration of resources. With the components included in the
vSphere suite, it is possible to provide an out-of-the-box solution to many com-
mon business challenges facing customers in their environments.


Virtualization Concepts 1-3
Managing Resources 4-7
High Availability and Disaster Recovery 8-12
Migration 13, 14
Other VMware Datacenter Software 15, 16
1. What is the benefit of virtualizing resources? (Choose two.)

a. More physical hardware is required.
b. Reduction of datacenter power footprint.
c. Fully utilize hardware resources that might be otherwise idle.
d. Business processes stay the same.

2. The operating system in a virtual machine requires that specialized drivers are
installed, after the server has been successfully booted.
a. True
b. False
3. Bare-Metal hypervisors are...

a. Hypervisors that run on top of another operating system.
b. Hypervisors that are only used by the cloud.
c. Hypervisors that run as their own operating system.
d. None of the above.
4. Once configured, vCenter provides which functions? (Choose all that apply.)
a. The ability to create virtual machines
b. High availability for the virtual machines
c. Dynamic resource scheduling of the virtual machines
d. Centralized management of resources
5. Clusters are only a container for resources.

a. True
b. False
6. Resource pools enable the configuration of which resources? (Choose all that
apply.)
a. Memory
b. High availability
c. Power consumption
d. Processor
7. Which of the following does VMware vSphere Distributed Resource Sched-

uler use to ensure optimal resource utilization?
a. ESXi hypervisor shares
b. VMware Converter
c. VMware vCenter Operations Manager
d. vMotion to migrate the VMs

Chapter 3: vSphere Components 43
8. Which vCenter function provides continuous availability?

a. VMware Fault Tolerance
b. VMware vMotion
c. VMware High Availability
d. VMware Application High Availability
9. What does VMware vSphere App HA use to monitor services?

a. VMware Tools heartbeats
b. VMware Hyperic
c. Network Pings
d. SNMP Traps
10. VMware Data Protection (VDP) can only replicate to other VDP nodes.
a. True
b. False
11. VMware Site Recovery Manager enables an administrator to configure a plan

that does that? (Choose all that apply.)
a. Enables failover of an entire site
b. Enables automated testing of failovers
c. Can be frequently tested
d. Monitors applications for failover
12. vSphere Replication is an ideal solution for an environment that has an RPO
of less than 15 minutes.
a. True
b. False
13. vSphere vMotion enables... (Choose two.)

a. A virtual machine to be migrated between ESXi hosts but downtime is
required.
b. Host maintenance to be performed during regular business hours.
c. Dynamic movement of virtual machines to ensure optimal performance
with DRS.
d. P2V migrations to occur.

14. P2V migrations are the process of converting physical servers to virtual ma-
chines.
a. True
b. False
15. The VMware vCenter Operations Management Suite provides metric roll-up
badges for each element in the environment. What are the three main badges?
a. Health
b. Efficiency
c. Risk
d. Availability
16. vCenter Configuration Manager provides which service(s) to a VMware

vSphere environment? (Choose all that apply.)
a. Configuration Management
b. Compliance Management
c. Dynamic resource scheduling
d. Centralized logging

Foundation Topics
Virtualization Concepts
It is important to understand the underlying concepts of virtualization before you
can understand how it can benefit the environment. In this section we discuss the
basic principles including virtual resources, virtual machines, and hypervisors. This
will build a framework to learn about the other technologies that increase the value
proposition of virtualization and also begin to show why it has become so popular.
Virtual Resources
In Chapter 1, “Datacenter Virtualization,” virtualization was introduced to provide
a background for why it is so popular in the business and technical worlds. VMware
uses a hypervisor to take the resources available in a physical server and abstract
them. These abstracted resources are known as virtual resources and are the core
building blocks of virtual machines. Figure 3-1 shows the four primary abstracted
resources.
CPU RAM Disk Network
Figure 3-1 Virtualized resources

These resources are the following:

■ Processor (CPU)
■ Memory (RAM)
■ Hard disk
■ Network
Each of these resources is discussed, providing background necessary to understand

how they are managed in a vSphere environment.
NOTE CPU and RAM are often combined and called compute resources in discus-
sions that are not specific to a particular resource.
Processor
The processor in a virtualized environment is used in the same way as in any com-
puter. It is the hardware that carries out the main processing for the VM. A virtual
CPU is a required component of a VM. The difference primarily lies in the fact that
the physical central processing unit (CPU) in the host server is managed by the hy-
pervisor, so that it can be shared amongst all the VMs running on the hypervisor.
VMs are also able to utilize more than a single CPU. As of vSphere 5.5, with the
use of virtual symmetric multiprocessing (SMP), each VM can use a maximum of 64
CPUs. The hypervisor therefore must be efficient at scheduling the access to CPU
resources to ensure peak performance of all running VMs.
In addition, ESXi hosts offer the ability to hot-add vCPUs to running VMs (as long
as the guest operating system supports it) to enable expansion of resources without
experiencing downtime.
Memory
The random access memory (RAM) in a virtualized environment is also used in the
same way as a physical computer. RAM is used to store the data that is currently
being used by the running operating system and its applications. It is volatile, so if
power is lost, what was in the memory is also lost. Virtual RAM is a required compo-
nent of a VM. The primary difference in the virtual environment is that the physical
RAM in the host server is abstracted so that it can be used by all the VMs running
on the hypervisor. As of vSphere 5.5, each VM can use up to 1TB of memory.

Due to how RAM is virtualized, the hypervisor is able to allocate more RAM to
VMs than exists in the physical host. This is done by creating a swap file on disk that
is used as RAM in the case that not enough physical RAM is available. This is called
over-commitment of resources. If resources are severely over-committed, swapping to
disk will occur more frequently and can have a significant impact on performance.
In addition, ESXi hosts offer the ability to hot-add RAM to running VMs (as long
as the guest operating system supports it) to enable expansion of resources without
experiencing downtime.
Hard Disk
Virtual machine hard disks (VMDKs) are used as permanent storage for VMware
VMs. In a virtualized environment, the “hard disk” is a set of files that exist on a
physical storage device—rather than an actual physical device. The device storing
the VMDKs could be either local or remote to the host. Commonly, VMDKs are
stored on a storage area network (SAN) to enable accessibility from multiple hosts.
The benefit to this is that it opens up a completely new way to manage servers due
to the fact that the disk is just a file on a datastore. With technologies such as thin
provisioning, snapshots, and linked clones, disks might contain only a small amount
of changed data, thus reducing the storage footprint for VMs and providing new
ways to provide data recovery in your environment.
Hard disks are not a hard requirement for a VM, although most VMs have them. It
is possible to boot a VM with no hard disk, in which case it must boot from another
device. As an example, this could be a boot from the network in which all files are
loaded into to memory in a stateless configuration.
In addition, you can also have multiple disks in a VM. This provides the ability to
separate operating system and data disks, as is common in many enterprise environ-
ments.
NOTE For more information on vSphere Networking, see Chapter 4, “vSphere

Storage.”
Network
In the virtual environment, networking is largely a software construct. The network
cards in a physical host are presented to the hypervisor and assigned as uplink ports
to a virtual switch (vSwitch). You then create portgroups on the switch to provide

networks that are utilized for functions in the environment. These portgroups can
be assigned for specific types of traffic, such as host management or VM.
This construct enables increased control over data security and performance over
the virtual networks. Functions such as Quality of Service (QoS) and network I/O
control are used here to prioritize traffic to ensure peak performance.
Network cards are a requirement for management of the hypervisor; however, VMs
do not have to have them. You can have multiple network cards in both hosts and
VMs.
NOTE For more information on vSphere networking, see Chapter 5, “VMware

vSphere Networking.”
Virtual Machines
Virtual machines are software-based containers that enable abstraction of a physical
server’s resources. As shown in Figure 3-2, without virtualization, the operating sys-
tem and applications consume the entire physical host, regardless of actual resource
utilization. Many times the application does not fully utilize the resources of this
host. This is not only a waste of resources, but also a waste of datacenter rack space
and—more importantly—power.
APP APP
OS
OS
1 3
Figure 3-2 Virtual machines visualized
By using VMs, you are able to abstract the physical resources and use them more
efficiently. This is because the hypervisor has control of the underlying resources.

Therefore, rather than having an idle server, many VMs can run and use the same
physical hardware simultaneously.
IMPORTANT Each VM is an independent and closed system in exactly the same

way a physical server is. In fact, for the most part, they are completely unaware that
they are running as a VM! The exception here is if they are running para-virtualized
drivers, meaning that there are functions that can be passed through to the physi-
cal hardware. Using para-virtualized drivers can provide a significant performance
improvement to a VM.
VMs are a core building block of the Software Defined Datacenter (SDDC) that en-
ables you to take existing physical servers and repurpose and consolidate them. VMs
are exactly the same thing as a physical server but use virtual hardware rather than
the underlying physical server hardware. Due to this, they enable all the benefits
provided by virtualization, such as high availability or dynamic resource scheduling.
This is shown in Figure 3-3.
APP
OS
APP
APP
OS
OS APP
APP
OS
OS
APP
OS
Figure 3-3 Hypervisor and consolidation

Hypervisors
The hypervisor is what makes virtualization possible. The hypervisor is the founda-
tional software that creates and runs VMs. There are two basic types of hypervisors:
■ Bare-metal hypervisors—Run directly on the physical hardware. These in-
clude VMware ESXi and Microsoft Hyper-V.
■ Hosted hypervisors—Run on top of another operating system.
The downside of hosted hypervisors is the added level of a host operating system
that significantly reduces the performance of the VMs because the hypervisor
doesn’t have ultimate control of the resources. These include VMware Workstation/
Fusion and Parallels Desktop.
Figure 3-4 shows the various types of hypervisors.
VM VM VM VM VM
VM VM VM VM VM Hypervisor
Hypervisor Host Operating System
Figure 3-4 Hypervisors
In this book, the discussion focuses primarily on VMware’s bare-metal hypervisor,

ESXi. Others are available as well, and this section discusses the following:
■ ESXi
■ Others
ESXi Hosts
VMware ESXi is the current implementation of VMware’s bare-metal hypervisor. It
was first released in 2001 and significantly enhanced the performance and reliability
of x86 virtualization. At the time, there was only hosted virtualization. In the years

since, it has seen several changes all that moved toward improving the current ESXi
architecture. In the current architecture, ESXi is a fraction of the size of previous
versions. More importantly, it dramatically improves management, security, deploy-
ment/configuration, and administration of the hypervisor.
ESXi hosts are a core component to the vSphere suite and enable virtualization to
be possible at the most fundamental level. They are the component on which the
VMs run and also perform many of the infrastructure management operations.
ESXi Hosts, however, are limited until it is a managed resource in VMware vCenter
Server. Once managed, a single ESXi server is no longer just another server and can
take advantage of pooled resources, high availability, and many other functions that
are fundamental to the vSphere suite. These functions and more are discussed later
in this book.
Others
Although VMware was the pioneer to bare-metal x86 virtualization, others are avail-
able in the market as well. A few of them are listed here for reference:
■ Microsoft Hyper-V—Released in 2008, Microsoft entered into the bare-
metal hypervisor market. Being integrated with Microsoft Windows, it allows
administrators who are familiar with Windows administration to easily take
advantage of virtualization in their environments.
■ Citrix XenServer—Xen originated as a research project at the University of
Cambridge, and a company, XenSource, was created to support the project.
The first public release of Xen occurred in 2003. XenSource was acquired by
Citrix in 2007.
■ KVM—Released in February 2007, kernel-based VM (KVM) is an open-
source hypervisor built in to the Linux kernel.
Managing Resources
Now that you understand the fundamentals of how virtualization works, we can
introduce the concepts of managing the virtual resources. Being able to manage
the virtual resources can be more complex than a physical environment with the
technologies in this section. VMware vCenter Server, clusters, resource pools, and
Distributed Resource Management all enable an administrator to manage the envi-
ronment. Many of the concepts are also unique to virtualized resources, effectively
shifting when and how specific functions are performed.

vCenter Server
VMware vCenter Server is the component of the vSphere suite that ties all of the
other components together. It provides a centralized platform for managing vSphere
environments, automating many complex tasks. After a host is added to the vCenter
Server inventory, vCenter provides the following functions:
■ Simple deployment—Templates, host profiles, cloning VMs, and vApps.
■ Centralized control and visibility—vSphere Web Client to manage the en-
vironment from any browser, vCenter Single Sign-On for authentication, cus-
tomized roles and permissions, and inventory searching.
■ Proactive optimization and availability—Resource management, Distrib-
uted Resource Scheduler (DRS), Distributed Power Management (DPM),
VMware high availability (HA), and vSphere App HA.
■ Management—vCenter Orchestrator to automate management tasks, perfor-
mance data, task and event logs, and API access.
Figure 3-5 shows vCenter and how it is central to managing a VMware environ-
ment.
VMware
vCenter Server
Automation Visibility
Unlocks the power Deep visibility into
of VMware vSphere every level of the
through proactive virtual infrastructure
management
Scalability
Scalable and Extensible
management platform
VMw
are v
Sphe
re
Figure 3-5 vCenter server

vCenter is a scalable management platform that allows up to a maximum of 1,000

ESXi hosts and 10,000 VMs per instance. It is available in both a Windows version
as well as a Linux appliance. This gives an administrator flexibility to choose the
platform used in his environment.
After an ESXi host is added to vCenter, you are presented with a view of the host.
This allows you to assess whether there are environmental or performance problems
in the infrastructure. vCenter can also automate the handling of common tasks or
unexpected failures.
In addition, vCenter can be augmented to enable other solutions to extend the func-
tionality provided out of the box. As an example, vCenter Site Recovery Manager
and VMware vCenter Converter can be added to extend the functionality provided
by the vCenter management platform.
Clusters
When adding a host to vCenter server, you must add it somewhere in the hierarchy.
The hierarchy in most cases will have at least one datacenter that contains either in-
dividually managed hosts or clusters of hosts.
Managed Hosts
A managed host is a host that is managed by vCenter but is not a part of a cluster.
Managed hosts can take advantage of central management (Single Sign-On access
control, logs, and basic host monitoring) and migration technologies (vMotion,
Storage vMotion, and Datastore Clusters). They cannot, however, participate in
functions such as VMware DRS, high availability, or virtual SAN (VSAN) activities.
Clusters of Hosts
A cluster of hosts enables vCenter to further abstract the resources of member ESXi
hosts into a pool of CPU, memory, and storage resources. As of vSphere 5.5, a single
cluster can contain up to 32 hosts. Once created, all the features of a cluster are
available to configure. These features include the following:
■ DRS
■ DPM
■ HA
■ Enhanced vMotion Compatibility (EVC)
■ VSAN

Resource Pools
Resource pools are an abstraction of CPU and memory resources. This abstraction
is used to provide a hierarchy and partitioning of the available resources into guar-
antees for VMs that are under the resource pool.
Resource pools can be configured on only standalone/managed hosts and DRS-
enabled clusters.
INTERESTING Although resource pools are an entity available to create on ESXi

hosts and DRS-enabled clusters, in fact the host or cluster is a resource pool as
well. These entities are known as the parent resource pool, and they are the ele-
ment that has all the available resources for that entity assigned to it.
Figure 3-6 shows the configuration of a resource pool.
Figure 3-6 Resource pools

Resource pools enable resources to be configured exactly as an administrator needs

to meet the business requirements, as shown in Figure 3-6. The customizable set-
tings that can be configured for a resource pool are as follows:
■ Reservations—Enable an administrator to assign a static amount of RAM
and/or CPU to the resource pool.
■ Shares—Enable an administrator to set a relative priority for the items using
the resources assigned. Shares are typically specified as High, Normal, or Low,
and these values specify share values with a 4:2:1 ratio, respectively. You can
also select Custom to assign a specific number of shares (that express a propor-
tional weight) to each VM.
■ Limit—Enable an administrator to limit the amount of resources that can be
consumed by the resource, even though free resources might be available. This
is commonly used to simulate resource starvation or to set expectations on the
level of performance of VMs in the resource pool.
■ Expandable reservations—The behavior of this resource pool if all resources
are in use. If the reservation is not expandable and there are insufficient re-
sources to fulfill the requirements of a child VM, an error will be presented
when attempting to power it on.
Distributed Resource Scheduler

VMware DRS is a feature of a cluster that can be enabled when an ESXi host is
added to a cluster. When added to a cluster, the physical host resources are then
added to the pool of available cluster resources. This enables all resources added to
the cluster to be aggregated and enables new functionality for the hosts. This func-
tionality can be split into two categories:
■ DRS scheduling
■ Distributed power management
DRS Scheduling Functions

When DRS is enabled, vCenter monitors the activity of the cluster and applies the
configured policies to the resources. This enables load balancing to be configured on
a cluster level and for VMs to be migrated using vMotion (discussed in detail later in
the chapter) between hosts in the cluster, as shown in Figure 3-7.

VM VM VM VM VM VMotion VM VM VM VM
Resource Pool
Physical Servers
Figure 3-7 DRS scheduling
While enabled, DRS monitors CPU and memory resources and generates recom-
mendations as to whether a migration to a different host will benefit the perfor-
mance. These recommendations are assigned a priority rating that can be compared
against a threshold set by the administrator for how conservative or aggressive DRS
is in automatically acting on the recommendations. DRS can be configured in three
ways:
■ Fully Automated—DRS generates recommendations for VM migrations and
initial placement and acts upon them in an automated manner; no administra-
tor intervention required.
■ Partially Automated—DRS automatically chooses where to initially power on
a VM but does not automatically act on recommendations for migrations. Ad-
ministrator intervention is required to apply the migration recommendations.
■ Manual—The administrator needs to choose where to power on a VM and
also apply any recommendations generated by DRS.
In any case, DRS is a powerful tool to ensure that there is automated monitoring for
workload contention in the environment, even if it is only acted on manually. It will
ensure that if resources are unbalanced, action will be taken.
Distributed Power Management Functions

VMware DPM is a function provided by DRS to reduce energy consumption dur-
ing off peak times. When enabled, DPM monitors the environment for activity on
the cluster and will consolidate VMs to as few ESXi hosts as are required to host

the workloads and still satisfy availability constraints. After the workloads have been
consolidated, DPM shut downs hosts into standby mode. This is shown in Figure
3-8.
VMotion VM VM VM VM VM VM VM VM
Resource Pool
Standby Host Server Power Optimized Servers
Figure 3-8 Distributed power management
DPM automates and optimizes energy management in the datacenter, so that an

administrator can cut the power costs during off peak times of resource utilization.
When utilization increases, DPM automatically restarts the ESXi hosts that have
been put into standby, thereby fulfilling the demand for resources.
High Availability and Disaster Recovery

After the environment is up and running, the focus must shift to keeping the en-
vironment available to end users, even if there is a failure of some kind. In this
section, you learn about VMware high availability, VMware fault tolerance (FT),
and VMware App HA, which enables quick recovery in the case of small failures of
individual entities. In addition, you learn about disaster recovery technologies, such
as VMware Data Protection, VMware Site Recovery Manager, and vSphere Replica-
tion, all of which enable larger disasters to be taken care of as quickly and efficiently
as possible.
VMware High Availability

VMware high availability (HA) is a component of VMware vSphere that provides an
easy-to-use and cost-effective solution for availability of VMs. It eliminates the need

for complex failover solutions that are tied to specific applications or operating sys-
tems. VMware HA as shown in Figure 3-9 provides the following benefits:
■ Helps to minimize unplanned downtime and service disruption.
■ Eliminates the need for dedicated standby hardware.
■ No additional software installation is needed on VMs.
■ Uniform availability across the virtual environment.
VM VM VM VM VM VM
Resource Pool
Figure 3-9 VMware HA failovers
After it is enabled, VMware HA continuously monitors the ESXi hosts in the cluster
for failures. This enables the environment to be proactively monitored for failures.
If a failure is detected, affected VMs are restarted on another one of the remaining
ESXi hosts in the cluster. VMware HA can detect full ESXi host hardware failures,
network isolations, and guest operating system failures.
VMware HA detects failures in the following ways:
■ Network isolation—If network connectivity fails on the management
network(s) that the agent uses to communicate between ESXi hosts, a host is
considered isolated.
■ Datastore heartbeats—If the network connectivity is detected as isolated,
the agent then checks to see whether the ESXi hosts are still accessible via the
datastores being used for heartbeats. This enables HA to determine if the host
is truly isolated or if the host has failed completely.

■ VMware Tools heartbeats—If enabled, HA monitors the VMware Tools

heartbeats from guest operating systems and can restart the VM if a heartbeat
is not received in a given amount of time.
Each cluster has the ability to protect up to 32 hosts. You can, however, have mul-
tiple clusters protected. This enables the environment to be proactively monitored
for failures to ensure the environment is available.
VMware Fault Tolerance

VMware FT is a feature that provides zero downtime, zero data loss, and continuous
availability for any application in the case of a server failure. Fault tolerance takes
high availability to the next level, using vLockstep technology. Once enabled on a
VM, the execution state is transferred across a dedicated network to a shadow VM,
as shown in Figure 3-10.
No Reboot
Seamless Cutover
VM VM VM Fault Tolerance VM VM Fault Tolerance VM VM VM
Figure 3-10 VMware FT failovers
If a server failure occurs, the shadow VM becomes active, providing instantaneous

recovery from a failed server. Due to vLockstep technology, which is operating at
the hypervisor layer, all actions are already transferred to the shadow VM. As a re-
sult, no agent is required for FT to function and it can be used with any VM, regard-
less of the operating system or applications running within the VM.

IMPORTANT As of vSphere 5.5, VMware FT is available only on VMs that have a

single virtual CPU.
VMware FT is used by organizations to

■ Completely eliminate service disruptions caused by server hardware failures.
■ Provide continuous availability for any application regardless of who devel-
oped the application.
■ Deliver uninterrupted service on demand that is simple and cost effective.
VMware Application High Availability

VMware application high availability (App HA) further extends and complements
the functionality of vSphere HA. It does this by leveraging VMware Hyperic to
monitor the applications being run in the VMs. This monitoring is done by an agent
installed to the VMs, as shown in Figure 3-11.
VMware vCenter Server
Hyperic App HA
Virtual Virtual
Appliance Appliance
VM VM VM VM
Agent Agent Agent Agent
VMware vSphere VMware vSphere
Figure 3-11 VMware App HA
The agent subsequently monitors the VMs reporting back to the Hyperic server,
which can act on changes configured in the configured App HA policy. App HA pol-
icies are configured by the administrator to match business requirements for the ap-
plications. App HA can therefore improve application uptime through the following:

■ App HA policies—Enable an administrator to configure a policy to monitor

and remediate an application based on business requirements.
■ Multiple recovery options—Enable an administrator to configure how the
application is recovered, such as a service restart, VM restart, or a combination
of the two.
■ Integration to vSphere HA—Enable an administrator to provide a full re-
covery solution for the application, VM, and host.
■ Uniform protection—App HA supports monitoring and protecting several
commonly used off-the-shelf applications. Not all applications are currently
supported for application level monitoring, unlike when you use standard HA
that fails over the entire VM.
■ Centralized management—App HA is visible from vCenter Server and
therefore is visible to the administrator through a familiar interface.
vSphere Data Protection

VMware vSphere Data Protection (VDP) is a backup and recovery solution with
built-in backup data replication. Traditionally, data backup and recovery solutions
have been built for a physical environment and are slow and complex. Often they
also write to a slow external device such as a tape drive. This adds to the complexi-
ties of designing a solution that meets business requirements.
VMware VDP is designed for the backup and restore of VMs in a vSphere environ-
ment. It is deployed as a virtual appliance powered by EMC Avamar. This provides
fast backup to disk with dependable recovery. Figure 3-12 shows a VDP environ-
ment.
VDP is fully integrated into vSphere and is easy to deploy and use. It can provide
both image-level backups as well as guest-level protection for applications in the en-
vironment. After deployment, you simply need to configure the policies that control
the data retention and schedules for backups of the VMs.
Once configured (assuming you have the appropriate license), it can also replicate
data between appliances or targets. You can replicate in the following ways:
■ One-to-one
■ One-to-many
■ Many-to-one

VDP
Virtual
VM VM VM Appliance
VMware vSphere
Figure 3-12 VMware VDP
In addition, you can replicate data to different targets, including these:

■ Another VDP appliance
■ EMC Data Domain systems
■ EMC Avamar grids
VMware Data Protection is a comprehensive solution for small and medium-sized

businesses to ensure that businesses are able to recover from failures and meet com-
pliance standards.
VMware vCenter Site Recovery Manager

VMware Site Recovery Manager (SRM) takes recovery to the next level. With VM-
ware FT or VMware HA, the focus is on the VMs and hosts. There is no solution
here if the failure is larger than a single host. This is where Site Recovery Manager
adds value into the environment, enabling an entire site to be failed over to a sec-
ondary site, as shown in Figure 3-13.

Site A (Primary) Site B (Recovery)
VMware Site Recovery VMware Site Recovery

vCenter Server Manager vCenter Server Manager
VMware vSphere VMware vSphere
Servers Servers
Storage Replication
Figure 3-13 VMware Site Recovery Manager
VMware SRM uses replication technology (Back end or vSphere Replication) to

replicate data between environments. After data is replicated, it provides the follow-
ing benefits to an environment:
■ Automated orchestration of a site failover and failback recovery plan for the
environment in case of a site failure
■ Nondisruptive testing of the recovery plan, which can be frequently tested if
needed by business objectives
■ Replicate storage with vSphere Replication rather than choosing from a third-
party replication technology
■ Protection of any virtualized application from a single solution that is applica-
tion and hardware agnostic
These benefits ensure fast and predictable recovery point and time objectives (RPOs
and RTOs) for an environment and can use existing underlying infrastructure if re-
quired.

vSphere Replication
vSphere Replication (VR) is a feature of the vSphere platform that allows an admin-
istrator to continually replicate a running VM to another location for disaster recov-
ery purposes. This enables rapid restoration of VMs in case a recovery is required. It
also enables flexible RPOs from 15 minutes all the way to 24 hours.
IMPORTANT If the recovery point objective is less than 15 minutes, vSphere Rep-
lication is not an appropriate solution.
After the images are taken, vSphere Replication can be used for restoring the images
through the vSphere web client or with a full disaster recovery product such as Site
Recovery manager. This is shown in Figure 3-14.
VM VM VM VM VM
VMware vSphere VMware vSphere VMware vSphere
VM VM VM VM
Figure 3-14 VMware Site Recovery Manager
vSphere Replication is a hypervisor-based replication technology that operates at

the VM disk level. This enables flexibility as it enables replication between any sup-
ported storage arrays. The storage technologies can differ as well, such as from a
Fibre Channel array to an iSCSI array.
With vSphere Replication in Sphere 5.5, you are able to use multiple point-in-time
snapshots that enable reversion to multiple previous states. This enables added flex-
ibility in creating a disaster recovery solution for the environment.
Migration
Virtualization also brings in technologies that enable movement of the virtualized
resources in the environment. VMware vMotion, VMware Storage vMotion, and

Physical to Virtual migrations are introduced in this section. These technologies en-
able unique new workflows in the environment, allowing maintenance windows in
the middle of the day to physical hosts and balancing of resources across the physical
infrastructure.
vMotion
VMware vMotion is a feature of vSphere that enables you to perform Live Migra-
tions of VMs between two separate physical ESXi hosts. vMotion is performed at
the hypervisor layer; therefore, it can be performed on almost any VM, regardless of
the guest operating system.
This process enables an ESXi host server to be cleared of running resources without
incurring any downtime at all for the running VM workloads. During the process,
the VM seamlessly transfers its running execution state and active memory over a
dedicated network link.
NOTE During a vMotion, the VM operating system does not realize that any-
thing is happening to it in the background. Thus, network connections proceed
normally and, other than a brief quiesce where generally no more than a single
ping is lost, the VM remains available to end users.
Figure 3-15 shows a vMotion.
VM
wa
re
ES
Xi
vM
oti
on
VM
wa
re
ES
Xi
Figure 3-15 VMware vMotion

vMotion relies on having identical (or very close to identical, most notably for the
CPU) underlying physical hardware on which the ESXi host is running. Thus, live
migrations are checked for host compatibility and allowed only between hosts that
can run the VM in the current execution state.
As a result, live migrations are not allowed between Intel and AMD CPUs (the ar-
chitecture is completely different) and are disallowed if networking and/or storage
elements are not shared between hosts.
IMPORTANT VMware has eased requirements between similar CPUs with a

technology called Enhanced vMotion Compatibility (EVC). VMware recommends
that a common EVC baseline be set on every cluster that will be performing vMo-
tions because it will help to prevent compatibility failures.
Once enabled, vMotion has the following benefits:

■ Automatic optimization of resources—VMware DRS requires vMotion to
balance workloads between servers. As a result, it is not effective unless vMo-
tion can be used to migrate VMs between all hosts in the cluster.
■ No downtime for hardware maintenance—Once enabled, vMotion can be
used with DRS to automatically evacuate a host of running VMs. This enables
you to perform hardware maintenance without scheduling downtime or dis-
rupting end users.
■ Movement of workloads—If a host is not performing well or if a host is fail-
ing, vMotion allows you to move the VMs to healthy servers.
Storage vMotion
VMware Storage vMotion expands on traditional vMotion technology by allowing a
VM to be live migrated not only to a different host, but also to a different datastore.
It enables an administrator to perform a migration of disks between datastores re-
gardless of the storage array or the class of storage, as shown in Figure 3-16.
Using storage vMotion, you can ensure that VMs are on the proper class of storage
for peak performance, as well as to utilize the underlying storage more efficiently.
In addition, much like with vMotion technology, VMs can be migrated off specific
datastores or storage arrays, so that maintenance can be performed on the under-
lying storage itself without needing to take down time on the VMs.

VM VM VM VM
VMware ESXi
Storage vMotion
Figure 3-16 VMware Storage vMotion
NOTE Much like with vMotion, the VMs themselves are unaware that a Storage
vMotion is occurring. As a result, there is no downtime to the VM when one oc-
curs.
Physical to Virtual
Physical-to-virtual migrations can also be performed with VMware vCenter
Converter. During this process, a physical Linux or Windows server, as well as
third-party image formats, can be converted into VMs that can run on VMware hy-
pervisors.
VMware Converter enables you to quickly migrate existing environments to virtual-
ized ones. This enables all applications in the environment to take advantage of the
benefits of virtualization, including the live migration and HA solutions offered by
VMware vCenter Server.

VMware Converter has the following benefits:

■ Performance and reliability—Quickly and reliably convert physical servers
into VMs, enabling you to adopt the benefits of virtualization rapidly.
■ Multiple conversion—Enable an administrator to convert multiple physical
servers at the same time, reducing the amount of time to adopt virtualization.
■ Centralized management—All operations can be monitored from the con-
sole simultaneously.
Other VMware Datacenter Software

In addition to the core features of the vSphere suite of products, technologies exist
that provide analytics, compliance, and configuration management to the environ-
ment. vCenter Operations Manager and vCenter Configuration Manager are tech-
nologies that enable an administrator to monitor the environment and report back
to the administrator.
vCenter Operations Manager

The VMware vCenter Operations Manager Suite (vC Ops) is an analytics tool that
gathers and analyzes performance information about the environment. It enables an
administrator to monitor for health, risk, and efficiency of the infrastructure to en-
able operations management to be automated.
Unlike with the default performance metrics collected with vCenter Server, vCen-
ter Operations Manager analyzes the data and takes monitoring to the next level.
This gives the administrator a deeper insight into the environment. It also enables
the administrator to proactively avoid performance issues. Unlike just having a pool
of metrics for each object in the environment, vCenter operations presents three
top-level metric roll-ups. This is shown in Figure 3-17. The three metrics that are
presented are
■ Health—Indicates immediate problems that might require your attention. It
helps to identify the current health of the system. It is a combination of work-
load, anomalies, and faults that are rolled up to present the metric.
■ Risk—Indicates a potential performance problem in the near future that may
affect the virtual environment. It is a combination of time remaining, capacity
remaining, and stress factors that are rolled up to present the metric.

■ Efficiency—Indicates opportunities to improve the performance or reduce the

cost of the environment. Efficiency accounts for wasted infrastructure as well
as the consolidation density in the environment. A large amount of wasted re-
sources will generate a poor efficiency score.
Figure 3-17 vCenter Operations Manager
vCenter Operations Manager uses these rolled-up metrics to enable a large number
of metrics to be presented in score form, dramatically simplifying traditional moni-
toring. By adding vCenter Operations Manager to the environment, the administra-
tor will add the following functions:
■ Automated roll-ups of key metrics into simple score metrics. This benefits the
administrator because there is no requirement to know what each and every
metric does to get an overview of the environment.
■ Calculation of a normal behavior threshold for every metric. This allows the
administrator to highlight abnormalities in the environment and take proactive
corrective actions.
■ Graphical representation of complex environmental stats.
■ Display of information about changes to the environment, such as adding
workload to an ESXi host and the resultant change to the health, risk, and ef-
ficiency of the host.

vCenter Configuration Manager

vCenter Configuration Manager (VCM) is a tool that checks configuration and
compliance management within an environment. You can use Configuration Man-
ager to automate the configuration management tasks across the environment.
Compliance can be verified not only in the virtualized environment, but also in
the physical Windows and Linux environments. It can be used to detect whatever
standard is required—whether it is an industry standard such as PCI or HIPPA—or
against internal standards in the environment.
Configuration Manager provides the following benefits to the environment:
■ Improved operational effectiveness by continually auditing configurations in
the infrastructure and being able to correct problems to ensure compliance
with a given standard configuration
■ Reduction of security threats due to changes in the environment
■ Minimized effort needed for configuration compliance in an automated
manner
■ Out-of-the-box compliance baselines to cover a broad range of standards that
are preconfigured
Summary
In this chapter, we discussed the vSphere components that not only make virtualiza-
tion a reality, but also transform the way in which businesses operate. With virtual-
ized resources, we are able to power VMs through the use of the hypervisor and
ESXi, in the case of VMware software.
After you have virtualized the environment, it can be managed by VMware vCenter
Server and its surrounding technologies. This gives an administrator a single-pane-
of-glass view of the environment and enables allocation of the resources dynamically
in the environment using clusters, resource pools, and DRS.
However, any environment needs to be properly protected from failure, and in the
availability section, you learned about HA, DRS, and App HA. These enable indi-
vidual resources to be protected. In addition to this, you learned about VDP, SRM,
and VR—technologies that enable a larger failure to be overcome or recovered from
as quickly as possible.
vMotion technology enables the administrator to take this a step further by provid-
ing the ability to live migrate VMs. This enables the administrator to proactively
perform maintenance or decommission a physical server, as required, without im-
pacting the availability of the virtual environment.

Finally, you learned about other technology that helps to monitor the environment.
VMware vCenter Operations Manager enables the administrator to monitor the
health, risk, and efficiency of the environment. On the other hand, vCenter Config-
uration Manager enables the administrator to monitor for compliance to a standard.
In any case, the technologies offered by VMware enable new ways to manage the
environment as well as let an administrator revolutionize how change control is
done in the environment. All of this is made possible by the basic constructs of virtu-
alization.

Key Topic Description Page

Element
List The key virtualized resources in the environment 46
List The types of hypervisors 50
List Types of management functions provided by vCenter Server 52
Figure 3-7 Functionality of DRS 56
List Types of HA failure detection scenarios 58
Figure 3-13 Functionality of SRM 63
List Benefits of vMotion 66
Figure 3-16 Storage vMotion conceptual diagram 67
List Types of metrics presented by vCenter operations manager 68


glossary.
Virtual Resources, Virtual Machine, Hypervisor, ESXi, vCenter Server, Cluster,
Resource Pool, DRS, DPM, HA, FT, App HA, VDP, SRM, VR, vMotion, Storage
vMotion, P2V, VC Ops, VCM
Review Questions
1. What is the benefit of virtualizing resources? (Choose two.)
a. Consolidation of resources.
b. Reduction of datacenter power footprint.
c. More physical hardware is required.
d. Business processes stay the same.
2. How does being virtualized affect the operating system in a VM? (Choose
two.)
a. The operating system is aware that it is virtualized.
b. The VM can be scheduled to use any CPU in the host.
c. The operating system is unaware that it is virtualized.
d. The VM must be tied to a physical CPU(s).
3. What are the benefits of using a bare-metal hypervisors? (Choose all that
apply.)
a. They run on top of another operating system.
b. They control the physical compute resources in the server.
c. They run as their own operating system.
d. The administrator configures specific amounts of compute resources for
the hypervisor.

4. Managing the environment with vCenter enables the usage of which addi-
tional vSphere features? (Choose all that apply.)
a. Templates
b. VMware HA
c. VMware DRS
d. VMs
5. Clusters are required for which features? (Choose all that apply.)
a. VMware HA
b. Resource pools
c. VMware DRS
d. vCenter Operations Manager
6. Resource pools enable the configuration of which resources? (Choose all that
apply.)
a. CPU affinity
c. Shares
d. Reservations
7. Which of the following does Distributed Power Management use to ensure

optimal resource utilization? (Choose all that apply.)
a. ESXi hypervisor shares
b. vMotion to migrate the VMs
c. VMware vCenter Operations Manager
d. Standby mode for ESXi
8. Which term best describes the type of availability that fault tolerance
provides?
a. Continuous availability
c. Replicated availability
d. Distributed availability

9. App HA can take which of the following actions if a monitored application

fails? (Choose all that apply.)
a. Restarting the ESXi host
b. Restarting the service in the guest operating system
c. Resetting the VM
d. Restarting the vCenter Server Service
10. VMware Data Protection plays which function in a virtualized environment?

a. Replication of data to the cloud
b. Backup and restore of VMs
c. Malware protection for VMs
d. Configuration management
11. VMware Site Recovery Manager is ideal for protecting against which type of
situations?
a. Power failure of a single host
b. A flood at the datacenter
c. Air conditioning failure in the datacenter
d. Installation of new ESXi servers to an existing environment
12. What is the minimum recovery point objective that vSphere Replication is
supported to use?
a. 5 minutes
b. 15 minutes
c. 30 minutes
d. 1 hour
13. vSphere vMotion is used for which purpose?

a. Changing the VM network from one port to another
b. VMware high availability failovers
c. Deploying templates from vCenter
d. Live migration of VMs between ESXi hosts

14. Which roll-up badge in the vCenter operations management suite provides a
view of how well the resources are being used in the environment?
a. Health
b. Efficiency
c. Risk
d. Availability
15. vCenter Configuration Manager provides which service(s) to the vCloud

Suite? (Choose all that apply.)
a. Centralized management of the environment
b. Automatic deployment of VMs
c. Template configurations
d. Compliance management

■ Physical Versus Virtual Storage—This section explains the difference

between physical and virtual storage.
■ Storage Types—This section covers the various storage types available to
vSphere.
■ Storage Protocols—In this section, the supported storage protocols avail-
able to vSphere are detailed.
■ Filesystem Types—In this section, you learn about the supported filesystem
types available to vSphere.
■ vSphere Datastores—This section demonstrates vSphere datastores and
how they provide storage.
■ vSphere Datastore Clusters—This section details vSphere datastore clus-
ters.
■ Virtual Machine Disk Types—In this section, you discover virtual machine
disk types.
■ Storage DRS and Storage I/O Control—In this section, you learn about
Storage DRS and Storage I/O control (SIOC).
■ Storage Profiles—This section discusses storage profiles and how they
function.
■ Other VMware Storage Technologies—In this section, you discover vari-
ous other vSphere storage technologies.
■ Availability—This section explains storage availability in vSphere.
■ Disaster Recovery—This section details vSphere technologies used to
recover from disaster.
This chapter covers a portion of the VCA-DCV objective 2.2.

CHAPTER 4
vSphere Storage
In this chapter, we discuss the formats and technologies used by vSphere to

present storage to virtual machines (VMs). Various filesystem types and storage
devices can be used to provide storage to vSphere. These technologies should
be designed as highly available because storage is an integral component of the
virtual datacenter. Several vSphere features ensure that VMs receive the perfor-
mance and capacity they require. All together, vSphere offers a robust storage
offering.



Virtual Versus Physical Storage 1
Storage Types 2
Storage Protocols 3
Filesystem Types 4
vSphere Datastores 5
vSphere Datastore Clusters 6
Virtual Machine Disk Types 7
Storage DRS and SIOC 8
Storage Policies 9
Other VMware Storage Types 10
Availability 11
Disaster Recovery 12
1. Which type of disk does not use spinning platters, increasing performance
drastically?
a. SCSI
b. SSD
c. SAS
d. SATA
2. Which type of storage is accessible via Ethernet?

a. Network Attached Storage (NAS)
b. SSD
c. Direct Attached Storage (DAS)
d. Distributed Storage
3. Which storage protocol is supported by vSphere?

a. FCoE
b. MTP
c. Firewire
d. CIFS

Chapter 4: vSphere Storage 79
4. Which filesystem was created by VMware for use with ESX?

a. NFS
b. VMFS
c. VMHD
d. VSHD
5. What is a datastore?
a. A shared storage device
b. A collection of disks aggregated together for virtual machine storage
c. A virtual hard disk drive
d. A logical container used to store virtual disks
6. What is a datastore cluster?

a. A shared storage device
b. A collection of disks aggregated together for virtual machine storage
c. A logical grouping of vSphere datastores
d. A logical container used to store virtual disks
7. Which type of virtual disk enables raw devices to be accessed by a virtual

machine?
a. RDM
b. VMDK
c. VFFS
d. RPM
8. Which vSphere feature ensures datastore latency does not surpass a defined
threshold?
a. NIOC
b. SDRS
c. SIOC
d. DRS

9. Which object(s) is storage policies applied to?

a. Virtual machines
b. Storage clusters
c. NFS
d. SIOC
10. Which VMware product is commonly used by SMB and branch offices to ag-
gregate local disk drives into a redundant shared datastore?
a. VSA
b. VVOLs
c. VAAI
d. VRFC
11. Which vSphere feature enables you to move a virtual machine’s data from one
datastore to another without interruption?
a. vMotion
b. Storage vMotion
c. Super vMotion
d. Storage replication
12. Which vSphere product can back up and restore entire virtual machines?
a. vSphere Data Watch
b. vCenter Backup Utility
c. vSphere Data Protection
d. vSphere Data Guard

Foundation Topics
Physical Versus Virtual Storage

Physical storage within a datacenter is the physical location at which data is stored
on storage media. All data within a datacenter resides on a physical storage device
connected to physical hosts via physical switches and cables. There are many types
of physical storage devices and filesystems. Depending on your business require-
ments, this storage type can consist of many forms. Physical storage can exist within
a host server itself or in specific, high-performance shared storage devices. Every
VM in a datacenter needs to reside on a physical storage device.
The most common storage media is a hard disk drive. The hard disk drive comes
in various formats, capacities, and speeds. Data is written and read to physical plat-
ters that spin. The rotational speed of these platters has been a major limiting factor
of storage performance for a long time. Solid state disks (SSDs) have changed this
by using memory instead of platters to store the bits of data. This has drastically
changed the physical storage landscape.
Virtual storage abstracts this physical storage and turns it into a format usable
for VM storage. Regardless of the type of physical storage device, as long as it’s
supported, vSphere makes it available in a standardized manner for VM storage.
vSphere is able to share storage devices across many vSphere hosts and enables the
seamless movement of VMs across them. VMware vSphere also greatly decreases
maintenance task impact by utilizing storage vMotion to move VMs away from stor-
age devices undergoing maintenance. vSphere allows environments to scale easily
and without disruption by using the same technology. vSphere offers a robust suite
of storage features and products that optimize and abstract physical storage, fully en-
abling the virtual datacenter. Figure 4-1 illustrates logically how hosts are connected
to storage devices in a storage area network (SAN) environment. The actual data at
the VMs on the hosts resides on the connected storage device.

vSphere Hosts Storage Device
SAN Switches
Fiber Channel Cable
Figure 4-1 A logical design of virtual and physical storage types
In this section, we discuss the types of storage used by the vSphere platform.
The types of storage are the following:
■ Direct attached storage
■ Network attached storage
■ Storage area network
Each storage type has different features and use cases, as well as different cost and
skills required to administer it.
Direct Attached Storage

Direct attached storage (DAS) is, as the name suggests, storage connected directly
to a host computer. This is most commonly found in the form of hard disk drives
(HDDs) plugged directly into a RAID controller on the host system. Other technol-
ogies allow direct connection of storage to a host system, but the principle remains
the same. DAS storage is presented as raw blocks of storage to the hypervisor; this
must then be formatted as VMFS to enable VM storage.
DAS storage was a very limiting factor of vSphere design until recently. DAS storage
cannot be shared amongst separate hosts. This, up until recently, removed the op-
tion to vMotion a VM from one host to another. In the vSphere 5.1 release, vSphere

enabled both a host vMotion and storage vMotion to occur at the same time, which
became known as Enhanced vMotion. This allows multiple hosts with nonshared
storage to operate in a similar manner to shared storage. The only real caveat to that
is that if a host fails, there is no way to move the VMs on that host to another server
until the server itself is available again.
Network Attached Storage

Network attached storage (NAS) is a file-level storage device connected to the net-
work. The storage device makes the storage available via ethernet and can connect
to many devices at once. NAS storage is common in datacenters that want to employ
a “one wire” type of architecture. This can greatly simplify the architecture of a
datacenter but can introduce some risk if not architected correctly. Many filesystems
are available on NAS storage, with the most common being NFS and CIFS.
Storage Area Network Array

A storage area network array is a block-based storage device connected to devices
via Fibre Channel (FC) or Ethernet. Fibre Channel uses traditional SCSI-based
commands to provide multiple hosts with block access to its storage in the form
of logical drives called logical units (LUs). In SCSI, each LU is assigned a refer-
ence number, referred to as a logical unit number (LUN). It has become common
practice for administrators to refer to the drives themselves as “LUNs.” The host
systems must then format the logical disks with whatever filesystem they require.
A Fibre Channel SAN is an independent network exclusive for block-based storage
and thus isolates storage traffic away from other forms of traffic. This might add
complexity to the architecture and support of an environment but reduces some risk
around shared devices for multiple purposes.
Storage Protocols
Numerous storage protocols are supported by vSphere, so the selection of protocol
is highly dependent on the environment and its design. The supported protocols are
as follows:
■ Fiber Channel
■ iSCSI
■ Fibre Channel over Ethernet (FCoE)
■ NFS
■ Local

Each protocol has strengths and weaknesses; no single protocol is better than an-
other. Choose what is appropriate for your environment and your business require-
ments.
Filesystem Types
A filesystem is the manner in which a host creates a layout on which to store data.
Two primary types of filesystem are available to vSphere: VMFS and NFS. VMware
created VMFS for the sole use of vSphere, beginning with ESX, to store VMs. NFS
is a distributed filesystem protocol in use by many types of operating systems and
hypervisors. There are advantages and design considerations to each.
NFS
NFS is a distributed filesystem protocol created and maintained on the network at-
tached storage device itself. The hypervisor does not control the filesystem of the
storage, nor does it need to. The storage device manages the filesystem and the
placement of data on it.
NFS provides many advantages in virtual datacenter design, such as the ability to
mount the filesystem on multiple operating system (OS) types concurrently. This
can aide in certain tasks such as backup and recovery. NFS has historically provided
larger datastores and more data visibility compared to older versions of VMFS.
VMFS has significantly closed that gap as of VMFS-5.
VMFS
Virtual Machine Filesystem (VMFS) was created by VMware to efficiently store and
access VMs. VMFS is written on block-based directly attached or shared storage de-
vices. VMFS is VMware’s cluster filesystem that allows multiple hosts to access the
same filesystem at the same time. It enables vSphere hosts to write and read from
the datastore simultaneously by using an advanced file-locking mechanism.
VMFS has undergone several version revisions, each with new functionality. Cur-
rently, VMFS is at version 5; today’s vSphere platform supports VMFS-3 and
VMFS-5. VMFS-3 should be used for compatibility only with ESXi hosts under ver-
sion 5.0. VMFS-5 allows for a unified 1MB block size as well as a 64TB maximum
datastore size.
vSphere Datastores
vSphere datastores are logical units of storage available for VMs, VM templates,
ISOs, and floppy images. Datastores utilize either the VMFS or NFS filesystem

dependent on the background storage. These datastores can be shared with multiple
hosts to a limit of 64 for VMFS and 256 for NFS (ESX5+), or they can be created
locally on a single host. The accepted best practice is to share datastores to all hosts
in a compute cluster. This enables VMs to move between all hosts easily and in-
creases availability in the event of a host failure. Figure 4-2 illustrates a datastore in
a vSphere environment.
Figure 4-2 This visualizes a datastore in the vSphere web console
vSphere Datastore Clusters

vSphere datastore clusters are a logical grouping of vSphere datastores. Datastores
of the same filesystem type can be aggregated together to create a single logical en-
tity. You cannot mix VMFS and NFS datastores in the same datastore cluster.
Datastore clusters allow for Storage DRS to be enabled, greatly reducing manage-
ment tasks for the storage layer. If more storage is required, a datastore can be
created and simply added to the cluster. This process allows the new datastore to
inherit all the settings of the cluster and allows Storage DRS to relocate VM storage
as necessary. Storage DRS is covered in more depth later in this chapter. Figure 4-3
shows a datastore cluster as it is viewed in the vSphere web client.

Figure 4-3 Demonstration of a datastore cluster in the vSphere web console
Virtual Machine Disk Types

vSphere VMs can be stored in two kinds of virtual disk type: virtual machine disk
(VMDK) or raw device mapping (RDM). VMDK is recommended for all workloads
unless a special use case requires an RDM. VMDK disk files are compatible with all
vSphere functions and features.
VMDK
A VMDK is essentially the hard disk drive of a VM. All data written and read for a
VM resides in the VMDK file. This VMDK file is stored on a vSphere datastore and
can be accessed by multiple hosts. A VMDK can be stored on any supported type of
storage device or filesystem. It is the standard building block of virtual storage.
A VMDK file can be provisioned in three ways:
■ Thin
■ Lazy-zeroed thick
■ Eager-zeroed thick
Thin-provisioned VMDKs allocate only the amount of space required for actual
data on a datastore. That means if you have a 20 GB C: drive but are using only
10 GB, it will consume only 10 GB of storage on the datastore. This type of

provisioning uses the least amount of storage and allows you to over-allocate datas-
tores. It does introduce some risk because if rapid data changes occur and you fill
up a datastore, the VMs on that datastore will cease to function. When new data is
written, the disk must first be grown, then a zero written, then the actual data. This
incurs the highest performance penalty of all the options but consumes the least
amount of space.
Lazy-zeroed thick provisioning reserves the entire 20 GB of space you have speci-
fied, but it does not write any data to the “empty” space. When new data is written,
a zero must be written first, then the actual data bit. This does incur a very small
performance penalty.
Eager-zeroed thick provisioning reserves the entire space allocated and proactively
writes zeroes for every block in the allocation. Eager-zeroed provisioning is recom-
mended for high I/O workloads such as SQL servers but consumes the most disk
space.
RDM
An RDM is a mechanism that enables vSphere VMs to access raw physical storage.
vSphere uses a VMFS volume with a mapping file to proxy commands to the raw
storage device; this allows for some vSphere features to be available (vMotion, HA)
while still offering direct access.
While not usually recommended, in certain use cases RDMs become advantageous.
Examples of this are when a cluster is created between a VM and a physical machine,
or when a specialized snapshot utility is required for backups.
Storage DRS and Storage I/O Control

Storage Distributed Resource System (SDRS) and Storage Input/Output Control
(SIOC) are advanced VMware features that aid in the management and performance
of VM storage. These features greatly reduce the amount of administrative time re-
quired to manage VM disk files and their performance.
SDRS
Storage DRS is a feature that can be enabled on datastore clusters and allows you
to select to have SDRS automatically perform storage actions for you or alert you
so that you can do them manually. Regardless of whether you intend to use SDRS
in automatic mode, it is useful to have it enabled. When enabled, it begins to collect
storage metrics that are otherwise unavailable or difficult to locate within vSphere.
This allows it to make more informed decisions if you decide to fully automate at a
later date. SDRS runs on a schedule that can be modified based on requirements. By

default, every 8 hours it analyzes the datastore cluster to evaluate whether any mi-
grations are required.
SDRS also provides affinity/anti-affinity rules for VMDKs; this means you can spec-
ify disk files that should reside together or apart. In general, it is best to leave them
together unless you have a specific reason to separate them.
Once enabled, many features are available to be configured for how to automate
your storage environment. Those options are as follows:
■ Free space threshold
■ Latency threshold
■ VM affinity/anti-affinity rules
The following figures demonstrate the available settings when configuring SDRS on
a datastore cluster. Figure 4-4 illustrates SDRS settings as displayed in the vSphere
web console.
Figure 4-4 SDRS in the vSphere web console
SIOC
Storage I/O Control enables storage traffic shaping and priority within a vSphere
environment. This is feature similar to Network I/O Control (NIOC) and compute
resource pools. A congestion threshold can be enabled to guarantee storage traf-
fic priority. A latency threshold can be configured to ensure a datastore is always
achieving below a configured millisecond response time. If the datastore goes above

that number, vSphere attempts to limit the I/O from all hosts to that particular
datastore in an effort to lessen the latency. This setting can also be set to a dynamic
percent threshold to better calculate the appropriate millisecond response type de-
pendent on the type of storage used.
Shares can be assigned to VM disk files that can give them higher or lower prior-
ity in the event of storage contention. As with all shares, administrators must be
careful when assigning these values because incorrect settings can have a profound
impact on the environment. SIOC will also evaluate all VM disk share values to as-
sist in VM priority in times of congestion. Figure 4-5 shows the settings available for
SIOC.
Figure 4-5 A datastore cluster in the vSphere web console
Storage Policies
Storage policies are policies that can be applied to VMs to ensure the VM is placed
in the correct storage. These policies can be derived using vSphere Storage API
for Storage Awareness (VASA), which polls the storage device for its capabilities, or
custom profiles can be created. After these profiles are created, they are assigned to
datastores. When deploying VMs, the storage policy is selected to ensure they are
placed in appropriate datastores.
Storage policies are also leveraged by other VMware products such as vCloud
Director. vCloud Director uses storage policies to ensure vApps are placed in the
proper datastores at the vSphere level. This also enables tiering of storage services
within a cloud environment. Figure 4-6 demonstrates selecting a storage policy dur-
ing VM creation to ensure it is placed in the correct datastore.

Figure 4-6 Selecting a VM storage policy at VM creation
Other vSphere Storage Technologies

VMware has many storage technologies for many different use cases. This creates a
large storage ecosystem to ensure VMs get the storage performance and availability
required. These technologies range from small business to enterprise-level function-
ality.
vSphere Storage Appliance

VSA is a software-based shared storage solution. An administrator installs an ap-
pliance provided by VMware into the vSphere environment. This appliance then
aggregates the local disk drives of the servers together into a logical datastore or
datastores and replicates data from one host to another. This ensures that in the
event of a host failure, the VM data exists on another host. This enables vSphere
high availability (HA) to properly bring the machine back online. VSA is limited in
the number of hosts it supports—either three or five. This is primarily targeted at
small or branch offices. It is useful for when no shared storage devices are available
or feasible for the environment.
Virtual SAN
A virtual storage area network is an object-based storage solution that aggregates
local disks into a policy-driven storage cluster. It is much more robust than the VSA
and offers support for up to 32 hosts in a cluster. Multiple types of storage profile

can be created to ensure performance that scales with a cluster. Multiple redundancy
options and cache percentages can be configured to provide very specific require-
ments to the storage platform. These profiles are then applied to VMs to ensure
they are placed in adequate storage for their requirements.
VSAN uses a combination of SSD and SATA/SAS hard drives to offer high per-
formance as well as capacity. VSAN is compatible with all vSphere features such as
vMotion, HA, and DRS. This product enables the functionality of a robust shared
storage platform utilizing the local capacity on the compute platform.
vSphere Flash Read Cache

vSphere Flash Read Cache (vFRC), formerly known as vFlash, allows a vSphere host
to use local SSDs as a high-performance read cache layer. vFRC pools together the
configured local SSDs in a cluster into one logical virtual flash resource. It then puts
a filesystem called the Virtual Flash File System (VFFS) on to that flash pool. What
this accomplishes is offloading read I/O from your storage devices. This enables
more storage bandwidth and cycles to process writes and non-cached reads and im-
proves performance.
Virtual Volumes
Virtual Volumes (VVOLs) is a technology that extends VM-centric storage to stor-
age devices. Currently, a storage administrator creates a LUN or volume on a stor-
age device and presents it to a host, which then consumes this as a datastore. After
the datastore is created, VMs are placed on it. VVOLs changes this to a different
process: the virtualization administrator creates a VM, and this sends a command
to the storage device to create the VM’s disk file. Included in these commands are
requirements for the capacity and performance of the disk. The storage device then
creates and places the disk according to the requirements. This greatly simplifies
storage management and is another way to ensure that the VMs policies are adhered
to.
Availability
Availability of storage is of paramount importance in a virtual datacenter. When
storage is offline, the VM is offline. In circumstances such as a host failure or net-
work device failure, the guest VM can in most cases be brought back online in very
short order. To minimize the impact of device or cable failure, vSphere enables
multiple paths to shared storage devices. A storage device failure has a much larger
impact because all VMs on that device will be unavailable until the storage device is
brought back online.

VMware has many technologies that can minimize the effect of storage failures and
storage maintenance. Storage vMotion enables VM disks to be moved from one
datastore to another while the VM is online. This can occur between VMFS and
NFS datastores regardless of the backend storage type. It truly frees the VM from a
physical storage device. Enhanced vMotion is a new vSphere feature available in the
web client that allows you to change a VM’s host and datastore at the same time. It is
vMotion and Storage vMotion combined into a single operation.
Products such as VMware Virtual SAN and the VSA are designed to make local
storage devices as redundant as shared storage devices. These technologies replicate
the data from local datastores across multiple hosts. This enables servers with only
local disks to provide a highly available platform for running VMs. Figure 4-7 dem-
onstrates the possible failures that can be sustained and still maintain data integrity
and VM uptime.
vSphere Hosts Storage Device
SAN Switches
Figure 4-7 Storage availability
Disaster Recovery
Disasters occur and take many forms—a datacenter fire, data corruption, or even ad-
ministrator error. Ensuring the integrity and availability of VM data is of paramount

importance in a datacenter. Without the data, there are no applications or services.

VMware uses various technologies to prevent data loss and minimize the impact of a
disaster.
vSphere Data Protection

vSphere Data Protection (VDP) creates image-level backups of VMs within your
environment. It essentially captures the entire VM entity and stores it in a specified
location. It accomplishes this by first taking a snapshot of a VM. A snapshot creates
a point-in-time capture of a VM. Essentially, the snapshot process stops the writing
of data to the VMDK file and instead writes all changes to a -delta file. This enables
VDP to make a clean copy of all the VM data. Snapshots themselves are not meant
as a backup mechanism. When a snapshot is removed, the hypervisor writes all the
data from the -delta file back into the VMDK and removes the -delta. The VDP
uses an appliance to offload the backup processing away from the VM. The appli-
ance also de-duplicates the data before storing it. De-duplication is a process that
replaces identical pieces of data with a pointer file that can save significant storage
space. vSphere Data Protection also uses a technology called Changed Block Track-
ing (CBT). This increases the efficiencies of backups by copying only the blocks that
have changed since the last backup. CBT can reduce backup times and increase the
number of VMs backed up within a specified timeframe.
In the event of a VM deletion or other similar disaster, the VM can be restored in
its entirety utilizing VDP. VDP can leverage CBT during restore times as well to
restore only blocks that have changed since the last backup and copy only those
blocks. This enables a highly efficient and timely VM restore.
File-level restores are also possible via VDP. Administrators can mount the backup
file and locate the file that needs recovery. File-level recovery is performed using the
VDP restore client.
VADP offers full backup and recovery for all operating systems supported by
vSphere. It does not require an in-guest agent; it is, however, recommended to have
vmtools installed.
Figure 4-8 illustrates a VM recovery using vSphere Data Recovery.

Virtual 1
machine
goes down X
VM VM VM VM VM VM
VMware ESXi & ESX VMware ESXi & ESX
Restore
VDP
VDP
2 Select VM 3 Restore
Images/ Virtual Machine
Files to Recover
Figure 4-8 A VM recovery using vSphere Data Recovery
vSphere Replication
vSphere Replication (VR) is a vSphere feature that copies VM information from
one location to another. These locations can be within the same cluster, a different
cluster, a different vCenter, or even a different datacenter. It is a continual copy of
a VMs state performed on a defined interval. As with VADP, it replicates only the
changes made since the last interval to reduce network bandwidth and increase copy
times. Administrators can choose up to a maximum of 24 points in time to be able to
recover to; this can be configured as 1 a day for 24 days or 1 an hour for 24 hours.
This is dependent on your requirements for recovery. vSphere replication is a per-
VM technology and can back up a maximum of 500 VMs. The number of VMs that
can be backed up might be significantly lower depending on your available network
bandwidth and storage capacity.
vSphere Replication operates at the hypervisor level and is the only true hypervisor-
level replication technology today. VR via the hypervisor copies VM data from the
host to a specified appliance in another location. The appliance caches the data until
all the data has been received. It then applies this to the VM disks of the protected
copy at the remote site.

In the event of a failure at the primary site, the VM’s protected copy can be recov-
ered in the remote virtual center via the vSphere web client. This cannot be per-
formed if the original VM is still accessible. This is a per-VM activity. VMware has
a product called Site Recovery Manager (SRM) that can automate the recovery of
VMs in the event of a disaster. SRM is an additional purchase and is not included
with vSphere Replication.
Figure 4-9 demonstrates vSphere replication between vSphere hosts.
Source VM Protected VM
VM VM
VMware ESXi VMware ESXi
VR
vSphere Replication Appliance
Figure 4-9 A VM replicated between hosts
Summary
In this chapter, we learned about the differences between physical and virtual stor-
age. Physical storage is required to store the actual data, while virtualization ab-
stracts this storage to make it highly available and easy to consume. Storage devices
come in many forms; virtualization simplifies this and greatly reduces management
complexity. Virtualization reduces the impact of physical storage maintenance on an
environment by being able to migrate workloads away from certain storage devices
without interruption of service.

The concept of datastores and how they are accessed was detailed. Datastores are
logical containers of either NFS- or VMFS-based storage. These can be assigned to
datastore clusters, which simplifies management. Several protocols are supported,
with each offering its own features. The selection of protocol is dependent on the
overall environment design. vSphere works well with all the supported options.
Storage DRS and SIOC showed how vSphere simplifies management and ensures
that VMs get the storage resources they require. Storage policies can be applied to
VMs to ensure their placement in appropriate datastores. These features combined
ensure the performance of VMs is within established thresholds.
Other VMware storage technologies were highlighted and their features explained.
VMware has a robust suite of storage technologies for all sizes of organizations and
required business use cases.
Disaster recovery options were examined; vSphere Data Protection and vSphere
replication offer robust disaster recovery capabilities and ease backup management
tasks. VDP and VR utilize hypervisor-level technologies to minimize the impact of
backup processes and speed up recovery times in the event of a disaster.


List Lists the three storage types 82
List Lists the four storage protocols 83
Section NFS 84
Section VMFS 84
List Lists the three types of disk provisioning 86
Section SDRS 87
Section SIOC 88
Paragraph Availability 92


glossary.
Solid State Disk (SSD), Network Attached Storage (NAS), Storage Array Network
(SAN), Fiber Channel (FC), Fiber Channel over Ethernet (FCoE), Internet Small
Computer System Interface (iSCSI), Network File System (NFS), Filesystem, Vir-
tual Machine Disk (VMDK), Raw Disk Mapping (RDM), Storage Distributed Re-
source Scheduler (SDRS), Storage I/O Control (SIOC)

These review questions validate you understand the topics covered in this chapter.
1. Which types of filesystems are supported by vSphere? (Choose all that apply.)
a. VMFS
b. CIFS
c. NFS
d. VFS
2. Which is NOT a supported storage protocol?

a. FCoE
b. iSCSI
c. FC
d. MTP
3. Which vSphere technology extends VM-centric storage to physical storage

devices?
a. VFRC
b. VSAN
c. VVOLs
d. VSA

4. Which is a virtual machine disk file created by VMware used to store virtual
machine data?
a. VFFS
b. VMDK
c. RDM
d. VMFS
5. Which storage feature enables automatic placement of virtual machine disk

files?
a. SIOC
b. NIOC
c. SDRS
d. DRS
6. Which feature limits host I/O to ensure datastore latency does not exceed a
specified threshold?
a. SIOC
b. NIOC
c. SDRS
d. DRS
7. Which storage protocol functions over a SAN?

a. CIFS
b. MPT
c. NFS
d. FC
8. What is the maximum number of intervals that can be stored by vSphere Data
Protection?
a. 24
b. 8
c. 16
d. 32

9. Which vSphere construct aggregates datastores together into a single logical

entity?
a. vSphere compute
b. VSA
c. VSAN
d. vSphere datastore cluster
10. On which vSphere object is a storage profile assigned?

a. Datastore
b. Datastore cluster
c. RDM
d. SDRS

■ Physical Versus Virtual Switches—This section explains the difference

between physical and virtual networking.
■ VMware Virtual Switches—This section describes VMware virtual switch-
es and their functionality.
■ VMware vSphere Networking Components—This section covers the
components of virtual networks and how they function.
This chapter covers a portion of the VCA-DCV objective 2.3.

CHAPTER 5
VMware vSphere Networking
Applications within a data center must communicate with each other and with
users in varying locations; the medium used for this communication is the
network. Devices, whether physical or virtual, must communicate with each
other in a standard way that all devices understand. All virtual network devices
must be able to communicate seamlessly on the network and do so using vir-
tual switches. There are multiple types and configurations of virtual switches;
this chapter covers these types and configurations to provide the foundational
knowledge required to understand vSphere networking.



Physical Versus Virtual Networking 1, 2
VMware Virtual Switches 3-6
VMware vSphere Networking Components 7, 8
1. Which is a feature that a physical network device can perform that vSphere
cannot currently perform?
a. Load balancing
b. Switching
c. Routing
d. Private VLANs
2. Which is a type of virtual switch available with vSphere?

a. Paravirtual standard switch
b. Consolidated virtual switch
c. Distributed virtual switch
d. Layer 3 switch
3. Which type of traffic shaping is available on a vSphere standard switch?

a. Inbound
b. Outbound
c. Ingress
d. Egress
4. Which load-balancing option is NOT available on a vSphere standard switch?

a. Route based on physical NIC load
b. Route based on source IP hash
c. Route based on source MAC hash
d. Route based on originating virtual port

Chapter 5: VMware vSphere Networking 103
5. Which types of discovery protocols are supported on a distributed virtual

switch? (Select two.)
a. Network Discovery Protocol (NDP)
b. Link Layer Discovery Protocol (LLDP)
c. Cisco Discovery Protocol (CDP)
d. IEEE Network Discovery Protocol (INDP)
6. Which feature on a distributed virtual switch allows you to connect ESXi hosts
to physical switches by using dynamic link aggregation?
a. Load balancing
b. LACP
c. Active/Active adapters
d. NetFlow
7. Which feature available on a distributed virtual switch allocates resources

based on shares?
a. Quality of service (QoS)
b. Network I/O control (NIOC)
c. Physical load-based teaming
d. NetFlow
8. Which feature available on a distributed virtual switch sends network traffic

information to a designated collector?
d. NetFlow

Foundation Topics
Physical Versus Virtual Networking

Physical networking uses multiple devices connected together via cables plugged
in to ports to facilitate the transmission of data from a source to a destination.
There must be a data path between all sources and destinations, including wireless
where components must be within physical range of each other. These connections
take many forms and can be very simple or extremely complex. Regardless of any
advancements made in virtual networking, a physical network is required. Virtual
networking offers many efficiencies over physical networking, but it still requires
physical cables connected to physical ports to move data between physical devices.
Advances in virtual networking are changing the way physical networks are designed
and implemented and will continue to do so in the future. Advances such as VM-
ware’s NSX offering will greatly change the way physical networks are configured
and managed—but a physical connection must still be made between devices.
In a traditional network world, an administrator would have to enter the data cen-
ter and cable a physical server to a physical switch every time a server was added
to the environment. That port would then need to be configured with the proper
configuration for the server to communicate on the network. This cabling and con-
figuration could take a fair amount of time; it is not uncommon for this to take days
to weeks. Virtual networking simplifies this procedure because the physical server
needs to be configured only once—that server being a vSphere host. After it’s cabled
and configured correctly, multiple virtual machines (VMs) can be provisioned or mi-
grated to the host and can immediately communicate on the network.
Virtual machines connect to virtual ports on a vSphere virtual switch using their vir-
tual network adapter. The virtual switch uses uplinks to transfer the network traffic
to the physical network; these uplinks are the physical network adapters of the host
server. For VMs on the same host connected to the same virtual switch, the physical
network is not required for them to communicate. This greatly increases the speed
at which applications can communicate. It also enables scenarios in which applica-
tions can communicate with themselves but not outside the host itself. This is par-
ticularly useful for development and testing requirements.
There are limitations on virtual switches—for example, they cannot perform routing
functions. This means that VMs that are in a separate network domains will need
to communicate with a router in the environment for the VMs to communicate.
There are also numerous limitations dependent on the type of virtual switch in use.
Features such as Network I/O control (NIOC) are not available on standard virtual
switches but are on distributed virtual switches (DVSes).

Virtual switches and the physical network must be designed in concert with each
other to ensure the proper functioning of communication between devices. Redun-
dancy, availability, and performance must be architected into the environment at
every level to ensure the network delivers what is required of it. Figure 5-1 shows an
example of a logical design for connecting the virtual network to the physical net-
work.
Network Core/
Router
Physical
Switch
vSphere
Host
Virtual
Switch
Virtual
Machines
Figure 5-1 A logical design of a virtual and physical network

VMware Virtual Switches

In this section, we discuss the various types of virtual switches available with the
vSphere platform.
The types of virtual switches are the following:
■ vSphere Standard Switch (VSS)
■ Distributed virtual switch
■ Third-party switches
Each switch type has differing feature sets and use cases, as well as different cost and
skills required to administer.
Virtual Standard Switch

The virtual standard switch (VSS) is a virtual switch located on each ESXi host. This
virtual switch uses the physical network cards in the host server as uplinks to the
physical network infrastructure. Internally, VMs connect to virtual ports using the
VM’s virtual network card. This is similar to the physical world, except the wires and
switch are created entirely in software. These switches are included on all versions of
vSphere. Figure 5-2 is a visualization of a VSS in the vSphere web client.
Figure 5-2 A virtual standard switch

Virtual standard switch settings include

■ Maximum Transmit Unit (MTU)—This specifies the maximum size of a
packet the switch can transmit or receive. By default, this is set to 1500, but
technologies such as VXLAN require an increase to 1600. Storage vendors
might also recommend an increase to an MTU size of 9000. MTU must be
an environment-wide setting because any device not configured correctly will
interrupt the traffic flow and cause problems.
■ Security
■ Promiscuous mode—Promiscuous mode allows network adapters con-
nected to this portgroup to view traffic not meant for them. Viewing this
traffic is useful for network monitoring software. In general, it is best to
leave this at the default value of Reject unless you are required to do
otherwise.
■ MAC address changes—All VMs are given a MAC address generated
by vSphere; this can be altered within a guest operating system. The
hypervisor will compare a VM’s configured MAC address versus what
is stored in the VM’s VMX file. If this setting is set to Reject, all those
machines using a MAC address that is not vSphere created will be unable
to communicate on the network. Microsoft NLB requires this to be set
to the Accept level as it uses a single MAC address for multiple servers.
Also, certain types of software licensing are tied to MAC addresses that
might require this setting to stay at the default Accept level.
■ Forged transmits—Forged transmits occur when a VM sends out traffic
identifying itself as someone else. vSphere will monitor outgoing packets
and compare the sender’s MAC address to the VM’s configured MAC
address. The hypervisor will then filter the communication based on the
configured setting, dropping those that do not match if the setting is
Reject. A common use case of this is nested ESXi hosts. Other services
might require this setting to remain at Accept, so refer to your company’s
security policy for all security guidelines.
■ Traffic shaping—Traffic shaping enables the configuration of the average
bandwidth, peak bandwidth, and burst size of network traffic. On a VSS, this
is available only on egress traffic, which means from the VM to the virtual
switch, not the other way. Both ingress and egress traffic can be shaped on a
DVS. Unless required, it is not recommended to enable traffic shaping.

■ Teaming and failover

■ Load balancing
■ Route based on IP hash—Route based on IP hash uses a calcula-
tion based on the IP address of the traffic to select an uplink. When
this path is selected, it will stay until an event occurs that forces a
recalculation of the path.
■ Route based on source MAC hash—Route based on MAC hash
uses a calculation based on the source MAC address of the traffic
to select an uplink. When this path is selected, it will stay until an
event occurs that forces a recalculation of the path.
■ Route based on originating virtual port—Route based on the
originating virtual port uses a calculation based on the original
virtual port of the traffic to select an uplink. When this path is
selected, it will stay until an event occurs that forces a recalculation
of the path. This is the default option.
■ Use explicit failover order—Explicit failover uses the specified
active and standby adapters. Traffic will flow over an active adapter
until it becomes unavailable, and then traffic will fail to the next
available adapter.
NOTE Load balancing must work in concert with the physical network topology.
Incorrect load-balancing settings at the vSphere layer can cause intermittent traffic
issues.
■ Network failure detection

■ Link status only—Link status failure detection fails the adapter
when it detects a link down. This could be due to an unplugged
network cable or a failed switch.
■ Beacon probing—Beacon probing requires three adapters to
periodically probe the network. This form of failure detection can
detect network problems that link status cannot but requires more
configuration.
■ Notify switches—This setting determines whether vSphere will notify
upstream switches of failure events.

■ Failback—When a network failure is resolved, this policy dictates

whether the traffic will be moved back to its original adapter or stay
where it currently is.
Distributed Virtual Switches

Distributed virtual switches (DVSes) are virtual switches similar in concept to a
virtual standard switch, but they offer much greater functionality. In the same man-
ner as a VSS, a DVS uses the physical uplinks of a vSphere host to uplink to the
network. The core difference between a VSS and a DVS is where the switch is cre-
ated and its configuration stored. A DVS is created and managed by vCenter Server.
vCenter then sends the configuration to each host that has been added to the DVS.
All changes to the DVS are done through vCenter—not individual hosts. This
greatly reduces the management complexity and reduces the risk of host networking
configuration not matching across a number of hosts.
As well as being centrally managed, many other features are offered in the DVS that
are absent in a VSS. A DVS offers NIOC, support for LACP, backup and restore
functions, NetFlow support, private VLANs (PVLAN), and Link Layer Detection
Protocol (LLDP). These features extend many commonly used enterprise network
principles in to the virtual world. Figure 5-3 visualizes a DVS in the vSphere web
client.
Figure 5-3 Adding a host to a DVS

Distributed virtual switch settings include

■ Properties
■ Version—On creation of a DVS, you can select the version of the DVS
itself, which enables compatibility with older infrastructures and can be
upgraded at a later date. Each version has specific features available to it.
■ Uplinks—Specify the number of uplinks for the DVS.
■ NIOC—Allows you to enable NIOC.
■ MTU—This specifies the maximum size of a packet the switch can
transmit or receive. By default, this is set to 1500. Technologies such as
VXLAN require an increase to 1600. Storage vendors might also recom-
mend an increase to an MTU of 9000. MTU must be an environment-
wide setting because any device not configured correctly will interrupt
the traffic flow and cause problems.
■ Discovery protocol
■ Type
■ Link Layer Discovery Protocol—An industry-standard network
protocol that provides information to switches about their neigh-
bors. It is useful when troubleshooting network issues because you
can easily tell which devices are connected to each other and where
possible problems might occur.
■ The Cisco Discovery Protocol (CDP)—This functions the same
as LLDP but is proprietary to Cisco devices. Cisco has allowed
VMware to use CDP to increase functionality within a Cisco envi-
ronment.
■ Operation—Operation specifies how the DVS will participate in the dis-
covery. Listen will gather information of all switches that are advertising.
Advertising will send information to switches but not listen. Both will
fully participate in the discovery process.
■ LACP—This setting enables the DVS to participate in link aggregation
groups that have been configured on physical switches.
■ Private VLAN—This enables the creation of private VLANs on the DVS.
This allows for highly secure networking constructs.
■ NetFlow—The DVS allows NetFlow information to be sent to a collector on
your network. Configuration of this setting coupled with a portgroup NetFlow
setting will send network information to the collector specified here.

■ Port mirroring—Port mirroring enables specific traffic to be relayed to an-

other source. This is useful for security or troubleshooting purposes.
■ Health Check—The DVS health check will perform a periodic check on the
DVS to ensure settings are correct.
Third-Party Virtual Switches

VMware offers extensibility to partners to create their own virtual switches to be
used with VMware. To date, only two partners have created custom virtual switches:
Cisco and IBM. Cisco created the Nexus 1000v, which replaces a VMware DVS
switch. The 1000v is managed in the same manner as all other Nexus network
devices. It gives network administrators in a Cisco environment end-to-end vis-
ibility to the network stack. However, the Cisco 1000v is not a common addition in
vSphere environments because it adds complexity to the vSphere environment and
requires a network administrator to manage it. Less common is the IBM 5000V vir-
tual switch, which offers the same common management plane as the Cisco 1000V
but for IBM environments.
All third-party switches replace a distributed virtual switch, not a vSphere standard
switch, thus requiring higher license requirements plus the additional cost charged
by the third party.
NOTE Third-party switch knowledge is not required for the VCA-DCV exam; it
is included for informational purposes.
vSphere Networking Components

Virtual networking has many components and configurations required to connect
VMs to the network. Components and configurations depend on the types of virtual
switches and the physical network topology in use. We will examine the configura-
tions of these components and describe potential use cases for each.
Portgroup
A portgroup is a construct created on the virtual switch that determines the network
connectivity for the attached VM. These portgroups are created per virtual switch
in the environment. A portgroup is a profile defining how a VM connects to the
network. These portgroups are then assigned to the VM’s virtual network card that

connects the VM to the network. The portgroup information, including case-sensi-

tive name, must be identical across all hosts on which the VM might exist. If a port-
group name is different across hosts, VMs will be unable to live migrate between
hosts. In large environments, the management of individual VSSes can be quite
cumbersome. This is due to the fact that each portgroup must be created on each in-
dividual vSwitch on each host in the environment. Distributed virtual switches ease
this administrative pain because the information is stored and configured via vCen-
ter server and then pushed to all hosts connected to the DVS.
Portgroups contain information such as the name of the portgroup and VLAN in-
formation. These options differ depending on the switch type on which they are cre-
ated, with the DVS having enhanced options. Figure 5-4 demonstrates how a VSS
with a management network portgroup looks in the vSphere web client.
Figure 5-4 A VSS with a management portgroup
Virtual standard switch VM options include the following:

■ vSwitch target
■ Existing
■ New

■ Connection settings
■ Network label—This name must match its equivalent portgroups on
other vSphere hosts.
■ VLAN ID (Optional)—If you are using VLAN trunking at the physical
uplink level, you must specify the VLAN ID number to tag on to traffic
for this portgroup. If you have a flat network without VLANs or specific
uplinks per VLAN, this setting is left as None.
Figure 5-5 shows a DVS portgroup and its settings in the vSphere web client.
Figure 5-5 A portgroup on a DVS
Distributed virtual switch VM portgroup options include

■ Name—This specifies the name of the portgroup and will be stored in a
central location, removing the need to ensure all configurations on all hosts
match.
■ Port binding
■ Static—Static binding binds a VM’s network adapter to a specific virtual
port and does not move it.
■ Dynamic—Dynamic binding binds a VM’s network adapter to the next
available port every time the machine connects to the DVS. This in-
cludes when a VM is powered off and on as well as high availability (HA)
events. (NOTE: This option has been deprecated and should no longer
be used.)

■ Ephemeral—This port type does not require a binding to a virtual port

and functions similarly to a VSS.
■ Network resource pool—The network resource pool is similar in function to
a compute resource pool. The default setting automatically detects the traffic
type and applies the appropriate NIOC settings. If user-defined resource pools
have been created, they can be assigned here.
■ Overriding of default policies—This enables the portgroup to override se-
lect settings inherited from the DVS.
■ Security
■ Promiscuous mode—Promiscuous mode allows network adapters con-
nected to this portgroup to view traffic not meant for them. Viewing this
traffic is useful for network monitoring software. In general, it is best to
leave this at the default value of Reject unless you are required to do oth-
erwise.
■ MAC address changes—All VMs are given a MAC address generated
by vSphere, which can be altered within a guest operating system. The
hypervisor will compare a VM’s configured MAC address versus what is
stored in the VM’s VMX file. If this setting is set to Reject, all those ma-
chines using a MAC address that is not vSphere created will be unable to
communicate on the network. Microsoft NLB requires this to be set to
the Accept level because it uses a single MAC address for multiple serv-
ers. Certain types of software licensing are tied to MAC addresses that
might require this setting to stay at the default Accept level, too.
■ Forged transmits—Forged transmits occur when a VM sends out traffic
identifying itself as someone else. vSphere will monitor outgoing packets
and compare the sender’s MAC address to the VM’s configured MAC
address. The hypervisor will then filter the communication based on the
configured setting, dropping those that do not match if the setting is
Reject. A common use case of this is nested ESXi hosts. Other services
might require this setting to remain at Accept, so refer to your company’s
security policy for all security guidelines.
■ Traffic shaping—Traffic shaping enables the configuration of the average
bandwidth, peak bandwidth, and burst size of network traffic. Both ingress and
egress traffic can be shaped on a DVS. Unless required, it is not recommended
to enable traffic shaping.

■ VLAN information
■ None—No traffic will be VLAN tagged.
■ VLAN—Allows you to specify a VLAN ID for the network traffic.
■ VLAN trunking—Enables you to specify a VLAN range for the traffic
on this portgroup.
■ Private—Allows you to configure private VLANs on the DVS.
■ Teaming and failover
■ Load balancing
■ Route based on IP hash—Route based on IP hash uses a calcula-
tion based on the IP address of the traffic to select an uplink. When
this path is selected, it will stay until an event occurs that forces a
recalculation of the path.
■ Route based on source MAC hash—Route based on MAC hash
uses a calculation based on the source MAC address of the traffic
to select an uplink. When this path is selected, it will stay until an
event occurs that forces a recalculation of the path.
■ Route based on originating virtual port—Route based on the
originating virtual port uses a calculation based on the original
virtual port of the traffic to select an uplink. When this path is
selected, it will stay until an event occurs that forces a recalculation
of the path.
■ Use explicit failover order—Explicit failover uses the speci-
fied active and standby adapters. Traffic will flow over and active
adapter until that adapter becomes unavailable; traffic will then fail
to the next available adapter.
■ Route based on physical NIC load—This load-balancing option
evaluates the load on all uplinks for the portgroup. If one link is
over 75% and used for a 30-second duration, vSphere begins to
move traffic to a lower-used link.
NOTE Load balancing must work in concert with the physical network topology.
Mismatched load-balancing settings at either the physical switch or the vSphere
layer can cause intermittent traffic issues.

■ Network failure detection

■ Link status only—Link status failure detection fails the adapter
when it detects a link down. This could be due to an unplugged
network cable or a failed switch.
■ Beacon probing—Beacon probing requires three adapters to
periodically probe the network. This form of failure detection can
detect network problems that link status cannot but requires more
configuration.
■ Notify switches—This setting determines whether vSphere will notify
upstream switches of failure events.
■ Failback—When a network failure is resolved, this policy dictates
whether the traffic will be moved back to its original adapter or stay
where it currently is.
■ Monitoring—NetFlow can be enabled for this portgroup; NetFlow settings
are configured on the DVS.
■ Traffic filtering and marking—This setting enables you to configure a num-
ber of traffic rules to apply to traffic for this portgroup.
■ Miscellaneous—Allows you to block all ports if required.
VMKernel Network Adapter

A VMKernel network adapter (vmk) is a special network interface for use by the
hypervisor itself. These vmk adapters are configured in a similar manner to a port-
group. They are created on a virtual switch in the environment with physical uplinks
to the network. These vmk adapters cannot be assigned to VMs because they are
for hypervisor use only. They have many of the same options as a portgroup and are
configured similarly, with a few specific options particular to them.
One option is to specify that this vmk adapter will be used for management traf-
fic. This management network vmk is how the host connects to vCenter server or
how an administrator connects to a host directly to manage it. This vmk adapter is
required for the manageability of a vSphere host in your environment and is created
upon install of a host by default.
Another common use of a vmk adapter is for vMotion. When you create a new vmk
adapter, you have the option to specify the type of traffic for which the adapter will
be used. When vMotion is selected, transferring of data required to move a running
VM from one host to another will occur over this link.

There are other uses for vmk adapters such as accessing network-attached storage
and for VMware fault tolerance (FT). You should follow VMware and vendor best
practices for the configuration of particular vmk adapter types. Figure 5-6 shows the
vmk adapter options in the vSphere web client.
Figure 5-6 A vMotion-enabled vmk network
Network Input/Output Control

Network Input/Output Control is a feature set that allows the prioritization and
modification of VM networking. NIOC can be used to prioritize network traffic
for certain VMs over others. It can also be used to limit the amount of bandwidth
a particular group of VMs has access to. NIOC uses network profiles assigned to
VM portgroups for its functionality. It is similar to the way resource pools for com-
pute function. Share values are used to calculate which VM has priority over which.
vSphere will detect the network traffic type and apply the appropriate NIOC profile
by default when NIOC is enabled. By default, the following profiles and priorities
are created:
■ NFS Traffic/Normal
■ Management Traffic/Normal
■ vMotion Traffic/Normal
■ Virtual SAN Traffic/Normal
■ vSphere Replication (VR) Traffic/Normal
■ iSCSI Traffic/Normal

■ Virtual Machine Traffic/High

■ Fault Tolerance (FT) Traffic/Normal
Custom share values and profiles also can be created; therefore, care must be taken
when using custom share values to ensure priorities are properly assigned. Custom
share values can be complex and introduce risk into the environment. Custom pro-
files assigned to portgroups will override the default NIOC profiles for that type of
traffic.
As with compute resource pools, NIOC priorities apply only in periods of conten-
tion. That is to say, that unless the host is constrained for network resources, all pro-
files will be treated equally. When contention occurs, machines with high priority
will be given greater access to network resources than those with normal priority.
All limits applied to network profiles will always be in place. It is highly recom-
mended to never use a limit unless you have an extremely good reason to do so.
Incorrectly implemented limits can have a very detrimental effect on the proper
functioning of applications. Figure 5-7 displays NIOC settings in the vSphere web
client.
Figure 5-7 NIOC management in the vSphere web client

Quality of Service
Quality of service is a way of prioritizing network traffic for a network environ-
ment. Whereas NIOC can control traffic only when it enters and leaves the vSphere
environment, QoS tags can function throughout an entire network environment as
long as it is configured end-to-end on all devices in the network. The QoS value is
configured on the same network profile NIOC uses but is not honored by vSphere.
vSphere uses the network share value and merely passes the tag along with the traf-
fic to the upstream infrastructure. QoS coupled with NIOC enables end-to-end pri-
oritization of network traffic throughout the datacenter.
QoS works on a rating scale with 0 at the highest and 7 at the lowest, which means
that traffic tagged with a 1 will have priority of traffic tagged with a 6. This is useful
in environments with applications that require very low latency, such as voice com-
munications.
NetFlow
NetFlow is an industry-standard feature that provides the ability to collect network
traffic that ingresses or egresses a network interface. Network flow information can
be sent from a DVS to a designated collector in the environment. This collector
can then view valuable network information, such as the source and destination of
traffic, and determine potential network bottlenecks and which protocols are in use.
This information is invaluable in network troubleshooting.
Summary
In this chapter, we learned about physical networking and how it differs from virtual
networking. Virtualization abstracts the physical network, simplifying network man-
agement and providing high availability. Virtualization is expanding more and more
into the physical network realm.
We discovered the concept of virtual switches and their features. The differences be-
tween a vSphere standard switch and distributed virtual switch were explored. Both
switches use portgroups to define network configurations that are then applied to
VM network adapters. Network traffic flows from virtual network adapters through
the virtual switches physical uplinks to the physical network, and vice versa.
Network I/O control and quality of service ensure VMs get the access to network
bandwidth that they require. NetFlow enables networking monitoring devices to
gather information about virtual network devices, offering end-to-end visibility.



Figure 5-1 Logical diagram of a networking design 105
List Types of virtual switch available with vSphere 107
List vSphere standard switch configuration options 108
List Distributed virtual switch configuration options 110
List vSphere Standard Switch portgroup configuration options 113
List Distributed virtual switch portgroup configuration options 113

glossary.
Network, Portgroup, Load Balancing, Link Aggregation Control Protocol (LACP),
NetFlow

These review questions validate you understand the topics covered in this chapter.
1. Which is a feature available on a DVS and not on a VSS?
a. Load balancing
b. VLAN tagging
c. Egress traffic shaping
d. Private VLANs

2. Which are two types of virtual switches available with vSphere? (Select two.)
a. Logical standard switch
b. Consolidated virtual switch
c. Distributed virtual switch
d. vSphere standard switch
3. Which type of traffic shaping is available on a distributed virtual switch but

not a vSphere standard switch?
a. Egress
b. Outbound
c. Ingress
d. NetFlow
4. Which load-balancing option is available only on a DVS?

a. Route based on physical NIC load
b. Route based on source IP hash
c. Route based on source MAC hash
d. Route based on originating virtual port
5. Which is an option to participate in Network Discovery?

a. Broadcast
b. Listen
c. Manage
d. Sync
6. Which feature on a distributed virtual switch allows you to tag traffic priority
in a manner respected by physical switches?
d. NetFlow

7. How many physical uplinks at minimum are required for beacon probing?
a. 1
b. 2
c. 3
d. 4
8. Which is a virtual network adapter reserved exclusively for vSphere hypervisor

use?
a. vmnic
b. vmk
c. nic
d. hvnic
9. Which type of link aggregation is available only on a version 5.1+ DVS?

a. ECP
b. Port bundling
c. SCP
d. LACP
10. Which is a construct that defines the connection parameters for VMs on a
virtual switch?
a. Portgroup
b. Network profile
c. vSwitch
d. vmk


■ Availability Challenge Solutions—This section discusses the technologies

that solve common availability challenges.
■ Management Challenge Solutions—This section discusses the technolo-
gies that solve common management challenges.
■ Scalability Challenge Solutions—This section discusses the technologies
that solve common scalability challenges.
■ Optimization Challenge Solutions—This section discusses the technolo-
gies that solve common optimization challenges.
■ SMB versus Enterprise Challenge Solutions—In this section, you learn
about the technologies that can aid both SMBs and enterprises in solving the
challenges that are unique to size of the environment.
This chapter covers VCA-DCV objective 3.1.

CHAPTER 6
Virtualization Solutions That

Solve Business Challenges
In this chapter, you solve the common business challenges introduced in Chap-
ter 2, “Business Challenges,” with the VMware technologies discussed in Chap-
ter 3,“vSphere Components”; Chapter 4, “vSphere Storage”; and Chapter 5,
“VMware vSphere Networking.” Virtualization enables revolutionary steps to
be taken to meet the common business challenges that customers face. In many
cases, the challenges are solved in not only a cost-efficient manner, but also in
ways that were never before possible.
For example, vMotion technology enables you to move virtual machines (VMs)
to a different host, with no downtime on the VM. This enables hardware main-
tenance to be done in the middle of the day because the ESXi hosts, when freed
from running VMs, are able to be taken offline in maintenance mode with no
impact to the business.
Virtualization and the corresponding abstraction of resources, truly can provide
solutions unique to your environment. Technology enabling low recovery time
objectives (RTOs) and recovery point objectives (RPOs) is no longer cost pro-
hibitively expensive whether you are a small or medium-sized business (SMB) or
a large enterprise, there is a solution that will benefit the business.



Availability Challenge Solutions 1, 2
Management Challenge Solutions 3, 4
Scalability Challenge Solutions 5, 6
Optimization Challenge Solutions 7, 8
1. Which of the following technologies will solve availability challenges? (Choose

all that apply.)
a. VMware SRM
b. vCenter Operations Manager
c. Dynamic Resource Scheduling
d. App HA
2. Which availability technology enables operating systems to be monitored by

VMware Tools?
a. VMware FT
b. VMware SRM
c. VMware HA
d. Application HA
3. Which of the following technologies solve management challenges?

a. vSphere HA
b. vCenter Server Templates
c. vCenter Operations Manager
d. vSphere DRS
4. Which management technology enables an administrator to reduce the power

and cooling requirements of the datacenter? (Choose all that apply.)
a. Virtual machines
b. vCenter Server
d. Application HA

Chapter 6: Virtualization Solutions That Solve Business Challenges 127
5. Which scalability technology enables an administrator to increase the number

of processors in a VM on-the-fly?
a. Resource pools
b. vSphere HA
c. Dynamic Resource Scheduling Allocations
d. Hot add
6. Which of the following technologies solve scalability challenges? (Choose all

that apply.)
a. VMware Storage Appliance
b. vSphere HA
c. VMware Distributed Resource Scheduler
d. vSphere Configuration Manager
7. Which of the following technologies can analyze the environment for optimi-
zation challenges?
a. vSphere HA
b. vCenter Server Operations Manager
c. VMware vMotion
d. vSphere Replication
8. Which optimization technology enables you to allocate more disk space than
is available in the host system?
a. Templates
b. Fast provisioning
d. Thin provisioning

Foundation Topics
Availability Challenge Solutions

Virtualization solves many of the availability challenges mentioned in Chapter 2,
“Business Challenges,” in unique ways. VMware provides solutions for all these chal-
lenges with the various products in the vCloud suite. In this section, we discuss the
solutions to these challenges.
Solutions to Availability Challenges

In Chapter 2, we discussed various types of challenges applicable to assessing avail-
ability of an environment. In this section, we discuss the pertinent technologies that
enable us to meet the different types of availability challenges directly. The types of
availability challenges are hardware, application, maintenance, and environmental.
The following describes the technologies that can be a factor for each:
■ Hardware—In normal operating configurations, a single component will be
at fault when a failure occurs. This could be as simple as a bad stick of RAM
or more complex such as a switch that services an entire rack failing. In many
cases, the actual physical redundancy will take care of it, such as routing the
traffic through another switch. But in the case of critical failures where an
ESXi host(s) fails, VMware HA and/or VMware FT are the features that will
react to the failure.
■ Application—When applications fail, it is a much different scenario because
VMware HA or VMware FT do not operate at this level. Thus, if the applica-
tion does not provide its own recovery options, VMware App HA can be an
option for monitoring of the application for failures.
■ Maintenance—Maintenance windows are completely revolutionized with
VMware technologies. To account for maintenance windows, Maintenance
Mode for clusters and VMware vMotion can enable VMs to be migrated to
different hosts preventing downtime from being needed for hardware mainte-
nance.
■ Environmental—Generally these are larger problems that impact a large
amount of the infrastructure. VMware SRM and vSphere Replication increase
environmental redundancy allowing VMs to be failed over or replicated to
a separate site. VMware SRM truly enables the concept of a Big Red Button
where an entire environment can be recovered to a separate site in the case of
a large-scale failure.

Solving Common Business Challenges

Given the different types of availability challenges, mapping the different technolo-
gies to different solutions is incredibly important. This enables you to provide an ef-
fective solution when presented with a common business challenge. For each of the
availability challenges listed in Chapter 2, a corresponding technology is discussed.
Hardware Availability
To solve common hardware availability problems VMware offers the following solu-
tions:
■ “My company needs to minimize the downtime caused by a host failing in the environ-
ment.”—vSphere HA will restart VMs in the case that there is a failed host in
the environment, therefore minimizing downtime.
■ “My company has a service level agreement that calls for a very quick response to a
host failing. Therefore, I have to be sure that VMs are up and running as quickly as
possible after a failure.”—vSphere HA automatically monitors for network con-
nectivity and datastore heartbeats. As a result, as soon as a failure occurs, VMs
will be failed over.
■ “I need to ensure that if a host loses access temporarily, appropriate action is taken.”—
vSphere HA monitors network availability for isolations and acts only if a host
is truly isolated and configured to failover in that condition.
Application Availability
To solve common application availability problems, VMware offers the following
solutions:
■ “We have an application that cannot be interrupted by hardware failures.”—vSphere
FT creates a shadow VM and will immediately switch over in the case that a
host failure is detected to ensure that there is no downtime for the application.
■ “I need to ensure that if my application stops responding, action is automatically taken
to try to resolve the problem.”—vSphere App HA monitors common business ap-
plications using VMware Hyperic application monitoring. It detects and cor-
rects problems with the service by restarting the service or server as configured
in the policy created by the administrator.

■ “My company requires that if an operating system stops responding for a period of
time, action is automatically taken to try to resolve the problem.”—vSphere HA –
VM Monitoring uses VMware Tools heartbeats to validate whether the guest
OS is running. If a heartbeat is not received within a defined period, the VM
will be restarted.
Maintenance
To solve common maintenance problems, VMware offers the following solutions:
■ “I need to be able to perform hardware maintenance at any time without impacting
the availability of services.”—vSphere vMotion enables VMs to be live migrated
to different ESXi hosts to allow hosts to be placed into maintenance mode and
taken offline.
■ “We need to avoid taking multiple servers down to apply hotfixes.”—vSphere vMo-
tion enables VMs to be live migrated to different ESXi hosts to allow hosts to
be placed into maintenance mode.
■ “My company needs to perform hardware updates at any time of the day without
impacting users who are working.”—vSphere vMotion enables VMs to be live
migrated to different ESXi hosts to allow hosts to be placed into maintenance
mode.
Environmental Availability
To solve common environmental availability problems, VMware offers the following
solutions:
■ “Our process to back up servers is difficult. It is hard to test for validity to ensure that
they are recoverable, and it is hard to restore an entire server quickly.”—VMware
Data Protection enables backups to be taken of entire VMs, which are easy to
test and recoverable very quickly.
■ “My company has mission-critical applications that must be available even if environ-
mental conditions cause problems at the datacenter. We would like a solution for this
that will also enable us to easily test the environment to ensure that any action that
is taken will be successful if we do have a disaster.”—VMware Site Recovery Man-
ager allows an administrator to configure a plan to failover an entire site to a
backup location. It accomplishes this by using backend array replication (or
vSphere Replication) to replicate changes that are applied based on the config-
ured recovery point and recovery time objectives.

■ “We would like to replicate server data to another site but have different types of
storage array at each site.”—vSphere Replication is a technology built in to the
ESXi architecture that enables VMs to be migrated between datastores regard-
less of the underlying storage technology.
Management Challenge Solutions

Virtualization solves many of the management challenges mentioned in Chapter 2
in unique ways. VMware provides solutions for all these challenges given the various
products in the vCloud suite. In this section, we discuss the solutions to these chal-
lenges.
Solutions to Management Challenges

In Chapter 2, we discussed the types of challenges that are applicable to assessing
manageability of an environment. In this section, we discuss the pertinent technolo-
gies that enable us to meet the different types of management challenges directly.
The types of challenges are datacenter, provisioning, and compliance. The following
describes the technologies that can be a factor for each:
■ Datacenter—Focusing on consolidation and centralized management of the
datacenter. VMware vCenter and VMs solve many of these challenges because
they enable consolidation and management to be greatly simplified.
■ Provisioning—Focuses on creation or migration of resources in the environ-
ment. VMware vCenter and VM cloning and templates enable an admin-
istrator to quickly provision standardized images for quick deployment of re-
sources.
■ Compliance—Focuses on adherence of the environment to a standard. VM-
ware Configuration Manager (VCM) enables an administrator to monitor the
configuration and compliance of resources in the environment.

Given the different types of management challenges, mapping the different tech-
nologies to different solutions is incredibly important. This enables you to provide
an effective solution when presented with a common business challenge. For each
of the management challenges listed in Chapter 2, a corresponding technology is
discussed.

Datacenter Management
To solve common datacenter management problems VMware offers the following
solutions:
■ “I am the only administrator for the environment, and I need a centralized way to
manage my datacenter.”—vCenter Server enables administrators to get a cen-
tralized view of the entire environment after it has been virtualized and allows
vSphere solutions such as high availability (HA) to be used.
■ “My datacenter is almost out of space. Power and cooling are maxed out, and it is diffi-
cult to introduce new hardware into the environment.”—VMs enable physical hosts
to be consolidated so that a single physical server can act as many, sharing the
resources of the server and consuming less physical space, power, and cooling
capacity in the datacenter.
■ “Our environment is complex, and we have many administrators around the world.
We need to be able to see everything around the world in a single pane of glass for
administrators.”—vCenter Server enables administrators to get a single-pane-
of-glass view of their virtualized environment. This includes being able to link
multiple vCenter servers together.
Provisioning Management
To solve common provisioning management problems, VMware offers the following
solutions:
■ “We have many physical servers running on really old hardware. This hardware is
out of warranty, and the applications are difficult to reinstall.”—Virtual Machines
and VMware Converter can be used to convert the older hosts into VMs, thus
keeping the applications available, even if they are legacy applications.
■ “I need to be able to provision a server from a standardized image that my organiza-
tion uses.”—vCenter Server can be used to create standardized templates that
can be used for deployment of resources.
■ “My company is growing quickly, and we need a way to rapidly deploy many VMs to
meet the demand.”—vCenter Server templates can be used to deploy many VMs
rapidly.

Compliance Management
To solve common compliance management problems, VMware offers the following
solutions:
■ “My company is in retail and needs to ensure we meet compliance regulations for credit
card information.”—vCenter Configuration Manager can be used to check com-
pliance of any regulation that has been configured and comes preloaded with
common baselines.
■ “We need to monitor our environment to make certain that our servers are running
the latest patches. If they are not, an administrator needs to be notified.”—vCenter
Configuration Manager can be used to check whether servers have specific
patches installed on them.
■ “I need to monitor an environment for unexpected changes to it. This will help to pre-
vent change control from being bypassed for critical servers”—vCenter Configura-
tion Manager can be used to check servers on a regular cadence to ensure that
they meet specific standards.
Scalability Challenges
Virtualization solves many of the scalability challenges mentioned in Chapter 2 in
unique ways. VMware provides solutions for all these challenges given the various
lenges.
Solutions to Scalability Challenges

In Chapter 2, we discussed the types of challenges that are applicable to assessing
the scalability of an environment. In this section, we discuss the pertinent technolo-
gies that enable you to meet the various types of scalability challenges directly. The
different types of challenges are infrastructure scalability and VM and application
scalability. The following describes the technologies that can be a factor for each:
■ Infrastructure scalability—The ability of the underlying infrastructure to
handle the needs of the servers in the infrastructure. The Virtual Storage Ap-
pliance (VSA), Virtual SAN (VSAN), and distributed virtual switches, enable
the infrastructure to be scaled to meet the needs of a growing virtualized envi-
ronment.
■ VM and application scalability—The ability to handle the growing load of
the servers in the environment. VMs, hot-add resources, VMotion, and Dy-
namic Resource Scheduling are technologies that enable the virtual servers to
be scaled out to meet the needs of the consumers.

Common Scalability Challenges

Given the different types of scalability challenges, mapping the different technolo-
gies to different solutions is incredibly important. This enables you to provide an ef-
fective solution when presented with a common business challenge. For each of the
scalability challenges listed in Chapter 2, a corresponding technology is discussed.
Infrastructure Scalability
To solve common infrastructure scalability problems, VMware offers the following
solutions:
■ “We are a small company and do not have a SAN. We still want to be able to take ad-
vantage of virtualization.”—VSA or VSAN will enable local storage to be used
and presented to the hosts as shared storage, so there is no requirement for
shared storage to take advantage of the benefits of virtualization.
■ “My company has the need to scale out the network bandwidth available to my hosts
without taking downtime on the VMs.”—DVSes enable additional network up-
links to be added, so that additional bandwidth can be provided to a switch.
VMware vMotion and DRS allow hosts to be evacuated and taken offline to
install additional NICs.
■ “I need to ensure that I can control the characteristics of the storage that is used to en-
sure that I have enough room to scale my environment appropriately to meet the needs
of my servers.”—VSAN enables the performance characteristics of the VM stor-
age to be configured dynamically with no downtime on the VMs or migration
to a different storage device.
Virtual Machine and Application Scalability

To solve common VM and application scalability problems, VMware offers the fol-
lowing solutions:
■ “My company has an application that requires dedicated servers and will be expanding
due to a merger with another company this year. We need to be able to scale out this
application quickly.”—VM templates can be used to quickly scale out an envi-
ronment as needed.
■ “We need the ability to add more CPU or RAM to the servers without having to take
the servers down.”—Hot-add capability for VMs is available to add both RAM
and CPU to VMs on-the-fly with no downtime.
■ “We need an automated system that will detect and correct issues with resource utili-
zation on our hosts or storage arrays.”—DRS, Storage DRS, and vMotion enable

automated monitoring and movement of resources to ensure performance is

optimal in the environment.
Virtualization solves many of the optimization challenges mentioned in Chapter 2 in
unique ways. VMware provides solutions for all of these challenges given the various
lenges.
Solutions to Optimization Challenges

In Chapter 2, we discussed different types of challenges that are applicable to as-
sessing the optimization of an environment. In this section, we discuss the pertinent
technologies that enable you to meet the various types of optimization challenges
directly. The types of challenges again are monitoring, efficiency, and performance.
The following describes the technologies that can be a factor for each:
■ Monitoring—Checking to see whether the environment is healthy, is ef-
ficiently operating, or there is risk that could impact the environment in the
future. vCenter Operations Manager enables proactive monitoring to be done
on the server to detect for the health, efficiency, and risk in an environment.
■ Efficiency—The ability to more effectively utilize resources to meet the
growing demand of the business without negatively impacting other servers.
vCenter Operations Manager can check for efficiency; however, technologies
such as storage profiles and thin provisioning can make the overall operation
and utilization of resources more efficient.
■ Performance—Ensuring that the environment is optimized for peak per-
formance as needed by the environment. vCenter Operations Manager pro-
actively can monitor performance to prevent issues before they happen. In
addition, technologies such as QoS on DVSes enable an environment to be
operating at peak performance.

Given the different types of optimization challenges, mapping the different tech-
nologies to different solutions is incredibly important. This enables you to provide
an effective solution when presented with a common business challenge. For each

of the optimization challenges listed in Chapter 2, a corresponding technology is

discussed.
Monitoring Optimization
To solve common monitoring optimization problems, VMware offers the following
solutions:
■ “We would like to be able to proactively monitor for performance bottlenecks in our
environment. Currently our environmental monitoring provides only a few basic
statistics that we can look at.”—vCenter Operations Manager enables an admin-
istrator to proactively monitor environment for bottlenecks and other perfor-
mance problems.
■ “My company needs to have detailed performance reports that can be used to trend our
environment.”—vCenter Operations Manager provides detailed performance
reports and heat maps of an environment to be analyzed by an administrator
or stored for historical purposes.
■ “I have several servers that are idle most of the time. I need a way to monitor them
and be able to assess whether consolidating them will affect performance.”—vCenter
Operations Manager provides projections of performance on a system and as a
result can predict what utilization will be on a system in the future given con-
straints to the resources.
Efficiency Optimization
To solve common efficiency optimization problems, VMware offers the following
solutions:
■ “We have several servers that have very little disk usage but were configured with a
large amount of disk space. It would be great to be able to minimize the impact this has
on the overall disk usage.”—Thin provisioning enables servers to not fully con-
sume the full physical storage space that has been allocated to the VM until it
is actually used for preserving disk space. These disks will grow as needed (up
to the maximum configured size), and an administrator therefore needs to be
vigilant about monitoring free space to ensure that errors don’t occur due to
low disk space.
■ “I need a way to detect whether I am efficiently using the available resources in my
environment.”—vCenter Operations Manager can detect whether resources are
being used efficiently in the environment.
■ “My company has several storage arrays that are used in the environment. Some of
these storage arrays are high performance and cost a lot more to operate than oth-

ers. Many times development users place VMs on the incorrect storage device, causing
degradation of performance for the array, which we correct by moving the VM to the
right class of storage. We need an automatic way to ensure that users can choose the ap-
propriate class of storage from the time of deployment.”—Storage profiles enable an
administrator to configure classes of storage. Thus, initial placement of VMs
can be ensured to be on the proper class of storage.
Performance Optimization
To solve common performance optimization problems, VMware offers the following
solutions:
■ “I need a way to proactively ensure that I have a healthy environment so I can get
optimal performance of my hosts.”—vCenter Operations Manager enables an ad-
ministrator to assess the resource utilization on hosts to ensure they are oper-
ating efficiently.
■ “We need a way to prioritize network traffic to ensure that our mission-critical VMs
get the peak performance possible.”—QoS for DVSes, and network I/O control
enables traffic to be prioritized for optimal performance of certain types of
traffic.
■ “My company needs to ensure that we can add resources to VMs in our environment
on-the-fly to address performance problems if they occur.”—Hot add resources can
be added to VMs to ensure that adequate resources are available to the server
to process incoming requests.
SMB Versus Enterprise Challenge Solutions

Virtualization solves many of the challenges mentioned in Chapter 2 for various
sizes of businesses in unique ways. VMware provides solutions for all these chal-
lenges given the products in the vCloud suite. In this section, we discuss the prod-
ucts that help to address the challenges that differ between SMB and enterprise
solutions.

Size of the Company

VMware takes into account scale while doing testing of vCloud Suite products.
Many times, the different solutions interact differently if you have a large environ-
ment. In fact, you might see problems with tasks that have previously been success-
ful previously, but fail due to scale.
VMware vCenter server enables an environment to be provided that scales to the
size of the company, providing a centralized management platform for a single-
pane-of-glass view of the entire virtualized environment. With vCenter, multiple
servers can be linked to provide a solution for an enterprise that has multiple vCen-
ter servers due to resource limits.
The following was the list of common challenges mentioned in Chapter 2. In this
section, we discuss the various technologies that need to be taken into account when
talking scale of an environment:
■ Amount of available bandwidth—Available bandwidth problems cannot be
mitigated other than by getting more of it. This being said, using a technol-
ogy such as QoS on a DVS can help to ensure that the traffic for high-priority
VMs is given preference. This might be a bigger problem for larger environ-
ments and as such ensure that it is taken into account when designing.
■ Backup strategy—vSphere Replication and vSphere Data Protection offer
different ways to replicate or backup the environment to take into account the
recovery time and recovery point objectives. They might not, however, provide
everything a large business would require over the needs of a small business.
Thus, a review of the configuration to ensure that either these products or
others meet the backup requirements of the environment is a must.
■ Availability strategy—Availability is a complex topic; however, VMware HA,
VMware FT, App HA, VMware SRM, and vSphere Replication all provide
varying levels of availability that can fit into the environment. Resource con-
straints, replication bandwidth, and/or site limitations might be bigger prob-
lems when looking at the size of the environment, and thus an appropriate
strategy needs to be designed to meet the requirements of the business.
■ Centralized management—VMware vCenter Server is the centralized man-
agement platform used to manage an environment. With larger businesses,
multiple vCenter servers can even be linked together to provide a single-pane-
of-glass view of the environment.

■ Compliance and configuration management—In the enterprise, software

such as vSphere Configuration Manager (VCM) are a must because with thou-
sands of hosts and VMs, it is next to impossible to quickly ensure that you are
in compliance.
Regardless of the technologies being used, these are the types of challenges that can
impact the design based on the size of the business. Enterprises can often have very
unique requirements that lead to larger expenses or a completely different solution
being required.
Summary
In this chapter, we discussed how the different virtualization technologies VMware
offers solve the challenges an administrator faces when designing and implementing
his environment. Technologies that meet the availability, management, scale, and op-
timization challenges of the environment were discussed.
Each VMware technology has a unique place in an environment that will revolution-
ize not only the way you operate in the environment, but also the way that you think
about the common business challenges.

Key Topic Element Description Page

List Solutions to challenges associated with availability 128
List Solutions to challenges associated with management 131
List Solutions to challenges associated with scalability 133
List Solutions to challenges associated with optimization 135

Key Technologies
List the technologies that solve the challenges for the Key Topics from this chapter,
and check your answers in Appendix A.
Availability Challenges, Management Challenges, Scalability Challenges,
Review Questions
1. Which of the following technologies will solve availability challenges? (Choose
all that apply.)
a. VMware FT
b. vCenter Operations Manager
c. VMware HA
d. Dynamic Resource Scheduling
2. Which availability technology enables an administrator to take into account a

power outage at a datacenter?
a. VMware FT
b. VMware SRM
c. VMware HA
d. Application HA
3. Which of the following technologies solves management challenges?

a. vSphere FT
b. Application HA
c. vCenter Configuration Manager
d. vSphere DRS

4. Which management technology enables centralized management of the data-

center?
a. vCenter Configuration Manager
b. vCenter Server
d. Virtual Machines
5. Which scalability technology enables an administrator to migrate VMs

on-the-fly with no downtime?
a. Resource pools
b. vMotion
c. vSphere HA
d. Hot add
6. Which of the following technologies solves scalability challenges?

a. VMware SRM
b. vSphere HA
c. Distributed Resource Scheduler
d. vSphere Configuration Manager
7. Which of the following technologies solves optimization challenges?

a. vSphere HA
b. Storage profiles
c. VMware vMotion
d. vSphere Replication
8. Which optimization technology enables you to monitor for risk in an environ-

ment?
a. Templates
b. Thin provisioning
d. vCenter Server Operations Manager


Answers to the “Do I Know
This Already?” Quizzes and
Chapter 1

1. A
2. B
3. A, B, C
4. C
5. A, D
6. B, D
7. A

1. B. ESXi and ESX are type 1 hypervisors. They operate on bare metal
servers abstracting physical hardware to be utilized by VMs. Type 2 hyper-
visors are installed on top of an operating system, not on the bare metal.
2. B. A VM is a logical container housing an operating system, data, and an
application.
3. A. A network is a core component of a datacenter. It, along with compute
and storage, form the three building blocks of a datacenter.
4. A. Transparent page sharing is the technology that removes duplicate
pages in memory and making more physical memory available to the host.
This enables memory oversubscription and higher consolidation ratios.
The host scans memory looking for duplicates; once found, it removes the
duplicate and replaces it with a pointer.

5. D. The Distributed Virtual Switch (DVS) includes advanced functionality not

available in the vSphere Standard Switch (VSS). These features include physi-
cal load-based load balancing, LACP support, and backup and recovery fea-
tures.
6. D. VMFS was created by VMware as a file system for block-based storage.
VMFS is highly efficient and designed specifically to store VMs.
7. B. A higher return on investment (ROI) means you are getting more out of
the money you are spending. VMware vSphere allows for more efficient use of
resources to ensure resources purchased are utilized and not wasted.
8. A. Virtual machine templates are preconfigured VMs that have been converted
or cloned to an entity marked as a template. These templates can then be rap-
idly cloned to deploy customized VMs.
9. D. Enhanced vMotion is a new vSphere feature available solely in the web
console. It enables a VM to migrate to a different host and datastore in the
same process.
10. A. Auto Deploy utilizes a predefined image that is deployed via the network to
a newly booted physical server. This physical server is then rapidly configured
to run as a vSphere host.
Chapter 2

1. C
2. B
3. B
4. A
5. A
6. A
7. D
8. B
9. C, D
10. A


1. B. Ensuring that there is minimal downtime in the environment is an avail-
ability challenge that must be considered by the administrator to ensure that
the needs of the business are properly met.
2. D. This is an availability challenge. Environmental protection almost always
boils down to how available the virtual machines are.
3. A. Centralized management and being able to see a view of resources in the
environment is a management challenge.
4. A. This is a management challenge. It is difficult to manage individual physical
hosts because power requirements will vary. In addition, the number of servers
allowed is limited by the available resources that can fit in a rack.
5. C. The ability to consume additional resources from the environment is a scal-
ability challenge.
6. D. Being able to scale the environment also applies to smaller sites that might
not have shared storage. Thus, this is a scalability challenge.
7. A. This is an optimization challenge. Efficiency review of the environment will
help to ensure that you are using resources effectively.
8. B. This is an optimization challenge. Health is a crucial area to look at when
ensuring the environment is operating effectively.
Chapter 3

1. B, C
2. B
3. C
4. B, C, D
5. B
6. A, D
7. D
8. A
9. B

10. B
11. A, B, C
12. B
13. B, C
14. A
15. A, B, C
16. A, B

1. A and B. Amongst many benefits that are provided by virtualization, it allows
you to consolidate resources by utilizing all the resources in the server. Rather
than having a low activity server sitting idle in the environment, many VMs
can be created, distributing the same resources amongst many servers. In addi-
tion, this in the end leads to reduced power consumption because fewer physi-
cal servers and their supporting physical infrastructure are required. Although
in the beginning, more physical hardware might be required to get the virtual
infrastructure up and running, consolidation can later be performed to decom-
mission older servers or to repurpose them, providing a significant gain in
the available resources. Business processes might also stay the same, however,
taking advantage of how virtualization can change businesses processes, due to
technologies such as HA and live migration.
2. B and C. The operating system running inside the VM is unaware that it is
virtualized and as a result can take advantage of many features such as vMo-
tion live migrations without impact to the running of the server. VMs are also
not tied to any physical CPU, unless you specifically dictate this configuration.
The ESXi hypervisor has a scheduler that allows the VM to run on any avail-
able CPU, preventing bottlenecks due to contention on a single CPU.
3. B and C. Bare-metal hypervisors run as their own operating system and have
full control of the resources available in the physical server. Although there are
options to limit the amount or compute seen by ESXi, these are advanced pa-
rameters and used only in very specialized circumstances, such as working with
VMware Support.
4. A, B, and C. Templates, HA, and DRS are all features enabled by using vCen-
ter Serer. VMs, however, are a construct enabled by ESXi and in fact can be
created without having a vCenter server at all.

5. A and C. VMware HA and VMware DRS are the only resources for which a
cluster is required. Resource pools can be configured on hosts that are not a
part of a cluster. vCenter Operations Manager also can monitor hosts that are
not a part of a cluster.
6. C and D. Resource pools enable shares and reservations to be configured for
CPU and memory resources. They do not enable the configuration of HA or
CPU affinity settings.
7. B and D. DPM uses vMotion to migrate the running VMs to different hosts;
then the host is put into standby mode, thereby preserving power in the envi-
ronment.
8. A. Fault tolerance provides continuous availability to VMs by replicating the
state of all activities on a VM to a secondary server. If a failure occurs, the sec-
ondary VM immediately takes over processing for that VM.
9. B and C. Application HA can either restart the failed service and/or restart the
guest operating system. It will not restart any ESXi host or vCenter services in
response to a service failure.
10. B. VMware Data Protection provides capabilities to quickly and efficiently
back up and restore VMs. It does not provide abilities to replicate to the cloud,
malware protection, or configuration management of the servers.
11. B and C. VMware SRM is used to protect the datacenter from a failure or
condition at a single site. Thus, if there is a flood or air conditioning failure
that is impacting an entire site, SRM can be used to failover mission-critical
servers to keep the business up and running. Power failure of a single host or
installation of new hosts would not require utilization of SRM.
12. B. The minimum recovery point objective for vSphere replication is 15 min-
utes. Any lower than this, as of vSphere 5.5, is not currently supported.
13. D. VMware vMotion is used for live migration of VMs between ESXi hosts.
None of the other technologies mentioned use VMotion.
14. B. vCenter Operations Manager provides the efficiency metric to allow ad-
ministrators to monitor for how efficiently the resources are being used. This
enables the administrator to proactively assess the environment to ensure re-
sources are not being wasted.
15. D. VMware Configuration Manager provides compliance management for the
environment, which can later be applied to ensure that all company or work-
load standards (ex. PCI DSS) are monitored and met.

Chapter 4

1. B
2. A
3. A
4. B
5. D
6. C
7. A
8. C
9. A
10. A
11. B
12. C

1. A and C. VMFS and NFS are the two supported filesystems by vSphere.
These filesystems form the datastores on which virtual machine files are
placed.
2. D. MTP is a protocol used by mobile devices to transfer media files. FCoE,
iSCSI, FC, and NFS are protocols supported by vSphere.
3. C. VVOLs communicate directly to storage devices and instruct them to write
VMDK files based on defined policies. These policies are defined on the VMs
and extend to the storage device itself creating VM-centric storage.
4. B. VMDKs are virtual machine disk files created by VMware and are the ba-
sic building blocks of virtual storage. All virtual machine data resides within a
VMDK file.
5. C. SDRS automates the placement of virtual machines on a datastore based on
configurable criteria. SDRS can also move virtual machines already placed in
incorrect datastores to more appropriate datastores.

6. A. SIOC monitors the latency of datastores when enabled to ensure they do

not surpass a defined threshold (default 30ms). When the threshold is violated,
SIOC begins limiting IO based on share values to return the datastore to an
acceptable latency.
7. D. FC (Fibre Channel) is a block-based protocol that functions over a SAN.
CIFS and NFS are file-based and operate via Ethernet. MPT is a mobile de-
vice media transfer protocol.
8. A 24 is the maximum number of intervals allowed. These intervals may be
configured as hours, days, or other required intervals.
9. D. A datastore cluster is an aggregation of datastores in a single logical entity.
Datastore clusters enable SDRS and greatly simplify storage management.
10. A. Storage profiles are assigned on datastores and define the datastores’ capa-
bilities. This profile can then be selected when deploying virtual machines to
ensure the virtual machine gets the required storage capabilities.
Chapter 5

1. C
2. C
3. D
4. A
5. B, C
6. B
7. B
8. D

1. D. Private VLANs are a feature that enables highly segmented and secured
networks and is available only on a DVS. All other features are common to
both the DVS and VSS.
2. C and D. vSphere standard switches (VSSes) and distributed virtual switches
(DVSes) are the two types of virtual switch available with vSphere.

3. C. Ingress traffic shaping is available only on a distributed virtual switch.

Egress traffic shaping is available on both the VSS and the DVS.
4. A. A DVS exclusively offers route based physical NIC load. A DVS and VSS
both offer route based on source IP hash, source MAC hash, and originating
virtual port.
5. B. Listen, Advertise, and Both are the options to participate in network discov-
ery.
6. A. Quality of service (QoS) is a standard tagging system that can be under-
stood by physical networking devices. This enables end-to-end control of VM
bandwidth priority.
7. C. At least three physical uplinks are required for beacon probing. This en-
sures there are enough uplinks listening for beacons to ensure the correct up-
link is taken offline in the event of a failure.
8. B. A vmk adapter is a virtual adapter reserved for use by the hypervisor. The
hypervisor uses a vmk for traffic such as management, vMotion, fault toler-
ance, and storage.
9. D. LACP (Link Aggregation Control Protocol) is available only on vSphere
version 5.1+ distributed virtual switches. LACP enables multiple physical links
to be aggregated together to form a single logical channel.
10. A. A portgroup defines connection parameters for VMs. It contains all the re-
quired information for the VM to communicate properly on the network.
Chapter 6

1. A, D
2. C
3. B
4. A, B
5. D
6. A, C
7. B
8. D


1. A and C. VMware FT and VMware HA both offer protection for different
types of availability challenges. vCenter Operations Manager and Dynamic
Resource Scheduling solve scalability and optimization challenges.
2. B. VMware Site Recovery Manager is the only technology that enables an en-
tire site to be failed over in the case of a power failure or natural disaster.
3. C. vCenter Configuration Manager is the only technology that solves a man-
agement challenge in the environment.
4. B. vCenter Server offers a view of the entire virtualized environment to enable
a single-pane-of-glass view.
5. B. VMotion offers live migration of VMs in the environment without down-
time.
6. C. VMware DRS is the only technology here that solves scalability challenges.
7. B. Storage profiles enable optimization of storage resources by enabling re-

sources to be automatically placed on appropriate classes of storage based on
an assigned policy.
8. D. vCenter Operations Manager enables an administrator to monitor an envi-
ronment for health, risk, and efficiency.


Glossary
Availability Availability refers to the amount of time an application is avail-

able to perform its function.
Availability challenges Challenges that relate to the uptime of the environ-
ment. These challenges could be due to hardware failure, environmental disas-
ter, or many other factors. The key to availability is that you need to ensure that
given the challenge, the system can take action to ensure proper recovery of the
environment within recovery time and recovery point objectives.
Availability challenge solutions VMware HA, VMware FT, App Ha, VM-
ware SRM, and vSphere Replication.
Clusters These provide logical aggregation of ESXi host resources that par-
ticipate in services which require shared resources, such as high availability.
Compute Compute is the collection of processing and memory resources
used to process guest application workloads.
ESXi A core component of vSphere, it is VMware’s bare-metal hypervisor.
FC Fiber Channel is a SCSI-based storage protocol.
FCoE Fiber Channel over Ethernet is Fiber Channel traffic encapsulated
within Ethernet.
Filesystem A filesystem is a layout created on disk used to control where bits
of data are stored and how they are accessed.
Hypervisor A hypervisor is a piece of software that abstracts the underlying
physical hardware and makes it available to isolated guest workloads. They are
the engine that makes virtualization possible. There are two types of hypervi-
sors, bare-metal and hosted, both of which provide the ability to create virtual
machines.
iSCSI Internet Small Computer System Interface is an IP-based storage pro-
tocol.

LACP The Link Aggregation Control Protocol is a network specification that en-
ables the bundling of several physical ports together to form a single logical channel.
Load balancing A network feature that enables the intelligent placement of traffic
for optimal performance.
Management Management is the means by which all resources are administered.
Management challenges Challenges that relate to the management of the envi-
ronment. These challenges could be as simple as ensuring there is centralized man-
agement of resources to being something as complex as consolidation of datacenter
resources to allow for further expansion.
Management challenge solutions VMware vCenter Server, VMotion, Storage
VMotion, templates, clones, vSphere Configuration Manager, and virtual machines.
NAS Network attached storage is a storage device connected to vSphere via Eth-
ernet and presents file-based storage.
NetFlow NetFlow is an industry-standard feature that provides the ability to col-
lect network traffic that ingresses or egresses a network interface.
Network A network is the means by which devices such as switches, routers, and
network adapters connect and communicate with the datacenter and outside world.
NFS Network File System is an industry-standard distributed filesystem protocol
available via Ethernet.
Optimization Optimization is making the most efficient use of available re-
sources.
Optimization challenges Challenges that relate to the effective utilization of the
resources in the datacenter. These challenges relate to monitoring the environment
for wasted space or to ensure that you are seeing peak performance of the resources.
Optimization challenge solutions vCenter Operations Manager, Dynamic Re-
source Scheduling (DRS), thin provisioning, storage profiles, and QoS for DVSes.
P2V migrations A physical server converted to a virtual machine.
Portgroup A portgroup is a profile defining how a VM connects to the network.
RDM A raw disk mapping is a disk type that gives virtual machines access to raw
storage.
Resource pools These provide a method of consolidating and managing compute
(CPU and memory) resources to enable reservations, shares, and limits to be set.
SAN Storage array network is a block-based network for storage traffic.

Glossary 155
Scalability Scalability is the ability of the platform to resize to accommodate more

demand for its services.
Scalability challenges Challenges that relate to the expansion of existing re-
sources to meet the demands of the business. These challenges mostly relate to the
available resources in either the physical infrastructure or the servers themselves.
Scalability challenge solutions Virtual machines, hot-add resources, Dynamic
Resource Scheduling (DRS), VSA, VSAN, and DVSes.
SDRS The Storage Distributed Resource Scheduler is a utility that ensures virtual
machine disk placement meets configured values for space and latency.
SIOC Storage I/O Control limits host I/O to ensure datastore latency is below a
specified latency after a defined threshold has been surpassed.
SMB versus enterprise challenges These are challenges that relate to the size of
the environment. Often small or medium-sized businesses have different require-
ments and challenges that would be different from that of a large enterprise. These
challenges are created by scale of available resources, such as the amount of band-
width available for the environment.
SSD Solid state disk is a physical disk type that replaces spinning platters with
memory. SSDs are incredibly fast but still somewhat expensive.
Storage vMotion Enables the live migration of virtual machine disks between dif-
ferent datastores.
Storage Storage is the location where data resides and is processed within a data-
center.
vCenter The core management platform for ESXi hosts, providing additional
functionality to the core installation.
vCenter Configuration Manager (VCM) Enables compliance and configuration
management in the environment.
vCenter Operations Manager (VC OPS) Enables a deep view into the health,
risk, and efficiency of the datacenter.
Virtual machine Virtual machines are servers that run on a hypervisor having ab-
stracted resources that are allocated and scheduled by the hypervisor.
Virtual resources Abstracted resources used by hypervisors. The four core virtual
resources are CPU, memory, storage, and network.
VMDK A virtual machine disk is a VMware-created file format used by virtual
machines to store and access data.

vMotion Enables live migration of running virtual machines between ESXi hosts.
VMware Data Protection (VDP) Enables the backup and restore of virtual ma-
chines.
VMware SRM Enables an entire site to be failed over according to a recovery
plan, in the case where you have a problem that spans an entire rack or datacenter.
vSphere App HA Provides application service monitoring via the VMware Hy-
peric agent and restarts the service or even the virtual machine to try to correct an
error.
vSphere DPM Part of VMware DRS that monitors ESXi host power utilization
and will put hosts into standby (and consequently take them out of standby when
needed) to save power during non-peak periods of utilization.
vSphere DRS A technology that uses vMotion to live migrate resources between
hosts in a cluster to ensure that resource utilization is optimized for the virtual ma-
chines.
vSphere FT Provides continuous availability to configured virtual machines, al-
lowing for a shadow virtual machine to take over in the case that the primary fails.
vSphere HA Monitors the hosts and virtual machines for failures and restarts
virtual machines to other hosts if one is detected. It uses both the network and the
storage infrastructure to ensure that servers are truly failed before it acts.
vSphere Replication (VR) Replicates virtual machine data between ESXi hosts to
enable quick recovery of virtual machines.

Try Safari Books Online FREE for 15 days
Get online access to Thousands of Books and Videos
FREE 15-DAY TRIAL + 15% OFF *

informit.com/safaritrial
Feed your brain

Gain unlimited access to thousands of books and videos about technology,
digital media and professional development from O’Reilly Media,
Addison-Wesley, Microsoft Press, Cisco Press, McGraw Hill, Wiley, WROX,
Prentice Hall, Que, Sams, Apress, Adobe Press and other top publishers.
See it, believe it
Watch hundreds of expert-led instructional videos on today’s hottest topics.
WAIT, THERE’S MORE!

Gain a competitive edge
Be first to learn about the newest technologies and subjects with Rough Cuts
pre-published manuscripts and new technology overviews in Short Cuts.
Accelerate your project
Copy and paste code, create smart searches that let you know when new
books about your favorite topics are available, and customize your library
with favorites, highlights, tags, notes, mash-ups and more.
* Available to new subscribers only. Discount applies to the Safari Library and is valid for first
12 consecutive monthly billing cycles. Safari Library is not available in all countries.

Pearson IT Certification
Articles & Chapters
THE LEADER IN IT CERTIFICATION LEARNING TOOLS
Blogs
Visit pearsonITcertification.com today to find: Books
Cert Flash Cards Online

IT CERTIFICATION EXAM information and guidance for
eBooks
Mobile Apps
Pearson is the official publisher of Cisco Press, IBM Press,
VMware Press and is a Platinum CompTIA Publishing Partner— Newsletters
CompTIA’s highest partnership accreditation

Podcasts
EXAM TIPS AND TRICKS from Pearson IT Certification’s Question of the Day
expert authors and industry experts, such as

Rough Cuts
• Mark Edward Soper – CompTIA
• David Prowse – CompTIA Short Cuts
• Wendell Odom – Cisco

Software Downloads
• Kevin Wallace – Cisco and CompTIA
• Shon Harris – Security Videos
• Thomas Erl – SOACP

CONNECT WITH PEARSON
IT CERTIFICATION
Be sure to create an account on
pearsonITcertification.com
and receive members-only
SPECIAL OFFERS – pearsonITcertification.com/promotions
offers and benefits
REGISTER your Pearson IT Certification products to access
additional online material and receive a coupon to be used
on your next purchase

Where Are the Companion
Content Files?
Register this digital version of
VCA-DCV Official Cert Guide
The VCA-DCV Official Cert Guide from VMware Press comes with three
practice exams. Each practice exam contains 50 questions, simulating the real exam,
and focuses on different objectives from the official VMware blueprint. Take each
practice exam in full and revisit the questions you answered incorrectly for remediation
and to identify the topic areas in which you need additional study support.
Follow the steps below to register this eBook and access the test questions:
1. Go to http://www.informit.com/register and log in or create a new account.
2. Enter this ISBN: 9780133854497
3. Answer the challenge question as proof of purchase.
4. Click on the “Access Bonus Content” link in the Registered Products section
of your account page to be taken to the page where your downloadable content
is available.
The Professional and Personal Technology Brands of Pearson

VCA-DCV Official Cert Guide

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

VCA-DCV Official Cert Guide

Uploaded by

Copyright:

Available Formats

From the Library of David Alfaro Cordova

From the Library of David Alfaro Cordova

For more information about VMware Press, please visit vmwarepress.com.

From the Library of David Alfaro Cordova

VMware® Press is a publishing alliance between Pearson and VMware,

Make sure to connect with us!

From the Library of David Alfaro Cordova

From the Library of David Alfaro Cordova

Upper Saddle River, NJ • Boston • Indianapolis • San Francisco

From the Library of David Alfaro Cordova

From the Library of David Alfaro Cordova

From the Library of David Alfaro Cordova

Chapter 2 Business Challenges 21

From the Library of David Alfaro Cordova

Chapter 3 vSphere Components 41

From the Library of David Alfaro Cordova

High Availability and Disaster Recovery 57

Chapter 4 vSphere Storage 77

From the Library of David Alfaro Cordova

Chapter 5 VMware vSphere Networking 101

Chapter 6 Virtualization Solutions That Solve Business Challenges 125

From the Library of David Alfaro Cordova

Management Challenge Solutions 131

From the Library of David Alfaro Cordova

Matthew (Matt) Vandenbeld is a dual VCDX, obtaining the certification in

From the Library of David Alfaro Cordova

From the Library of David Alfaro Cordova

From the Library of David Alfaro Cordova

Doug Baer is an infrastructure architect on the Hands-on Labs team at VMware.

From the Library of David Alfaro Cordova

From the Library of David Alfaro Cordova

From the Library of David Alfaro Cordova

Who Should Read This Book

Goals and Methods

From the Library of David Alfaro Cordova

How to Use This Book

From the Library of David Alfaro Cordova

Access to Practice Exams and Book Content Updates

2. Enter the ISBN: 9780133854497.

3. Answer the challenge question as proof of purchase.

From the Library of David Alfaro Cordova

■ What Is Virtualization?—This section explains what virtualization is and

From the Library of David Alfaro Cordova

The majority of server workloads now run on a virtualization platform. VM-

“Do I Know This Already?” Quiz

From the Library of David Alfaro Cordova

1. What type of hypervisor is ESXi?

2. What is a virtual machine?

3. Datacenters are comprised of which component(s)? (Choose all that apply.)

4. What vSphere Compute feature enables vCPUs to be scheduled efficiently?

From the Library of David Alfaro Cordova

6. What types of file systems are supported as virtual machine datastores?

7. What is a common business reason to virtualize physical server workloads?

From the Library of David Alfaro Cordova

From the Library of David Alfaro Cordova

VMware vCenter Server

VMware vSphere VMware vSphere VMware vSphere

Figure 1-1 The abstraction performed by the hypervisor

From the Library of David Alfaro Cordova

NOTE Compute is covered in depth in Chapter 3, “vSphere Components.”

From the Library of David Alfaro Cordova

NOTE Network is covered in depth in Chapter 5, “VMware vSphere Networking.”

From the Library of David Alfaro Cordova

■ RAID5 Parity—Requires a minimum of three drives and is a common balance