VMware Vsphere Design Essentials - Sample Chapter
VMware Vsphere Design Essentials - Sample Chapter
pl
e
P r o f e s s i o n a l
Puthiyavan Udayakumar
Sa
By the end of the book, you will have also learned about
the major factors influencing vCloud needs and have
finalized the requirements for designing a vCloud.
P U B L I S H I N G
VMware vSphere
Design Essentials
ee
E x p e r t i s e
D i s t i l l e d
VMware vSphere
Design Essentials
Unleash the performance, availability, and workload efficiency
of your virtual data center using this fast-paced guide
Puthiyavan Udayakumar
P U B L I S H I N G
Preface
This book provides the framework and methodologies that need to be applied
during the initial, planning, analysis, and design phase of the VMware vSphere
implementation in data centers.
This book will provide readers with the design fundamentals essential for a
VMware engineer and architect, even before they implement VMware vSphere
in their data centers.
After reading this book from cover to cover, readers will achieve the following:
The essential skills required to design VMware vSphere network and storage
Readers will gain a great deal of advantage from this book. They will learn how to
design a virtual data center with the given requirement and the key components that
will yield great success to business and IT. They will also learn how to convert the
industry standards to customer-specific by following the golden principle of the best
practices recommended by VMware.
Upon finishing this book, readers will have the skills required to design, build,
deploy, and manage VMs in a VMware vSphere data center with VMware ESXi
and VMware vCenter Server.
Preface
Essentials of
VMware vSphere
Thanks for choosing VMware vSphere Design Essentials, your companion in learning
the fundamentals of designing VMware vSphere. We understand your mission to
learn, apply, and reap the benefits of virtualization and to design VMware vSphere
to its fullest extent using this book. Let's get started learning about VMware vSphere
and its essentials.
VMware vSphere is a product developed by VMware, Inc. The company is located
at Palo Alto, California, USA, and was started in 1998. It offers virtualization,
cloud software, and services. VMware vSphere is a product that aims to provide
x86 virtualization to renovate datacenters into streamlined cloud computing
infrastructures, which in turn enables IT organizations to deliver consistent and
elastic IT services.
vSphere virtualizes physical hardware and converts one single physical system into
multiple VMs; thereby VMware vSphere reduces the space, cost, and complexity
of managing systems in a datacenter. This book is not only for VMware architects
but also for people who use vSphere on a daily basis. This book will help you
understand how vSphere is designed and will help you to design your virtual
infrastructure using VMware vSphere to its best potential. Also this book will help
you to improve your skills; you will become well versed in designing best practices.
Designing VMware vSphere infrastructure will be a multipart subject. In this
chapter, we'll provide an introduction to the VMware vSphere landscape, the design
of vSphere itself, challenges and obstacles that were caused by virtual infrastructure
due to poor design, and the way to overcome those challenges with structured
principles and processes.
[1]
Virtualizing networks
Virtualizing storage
Automated monitoring
Description
ESXi
Management
layer
[2]
Chapter 1
Components
Description
vCompute
vNetwork
This links VMs to one other within ESXi host, establish the
communication between VMs virtual NIC to the physical network, and
along with that also provides a communication chancel services for
VMkernel services (including NFS, iSCSI, vMotion)
We will see more information on designing vNetwork in the upcoming
chapters.
vStorage
Database
This acts a data management point to organize all the configuration data
for the VMware vSphere infrastructure.
We will see more information on designing databases in upcoming
sections in this chapter.
[3]
[4]
Chapter 1
When you release a design to the customer, keep in mind that the design must have
the following principles:
Without the preceding principles, I wouldn't recommend you to release your design
to anyone even for peer review.
For every design, irrespective of the product that you are about to design, try the
following approach; it should work well but if required I would recommend you
make changes to the approach.
The following approach is called PPP, which will focus on people's requirements, the
product's capacity, and the process that helps to bridge the gap between the product
capacity and people requirements:
The preceding diagram illustrates three entities that should be considered while
designing VMware vSphere infrastructure.
[5]
In the end, using this unified framework will aid you in getting rid of any known
risks and its implications.
Functional requirements should be meaningful; while designing, please make sure
there is a meaning to your design. Selecting VMware vSphere from other competitors
should not be a random pick, you should always list the benefits of VMware
vSphere. Some of them are as follows:
[6]
Chapter 1
The PPP Framework helps you to get started with requirements gathering, design
vision, business architecture, infrastructure architecture, opportunities and solutions,
migration planning, fixing the tone for implementing and design governance. The
following table illustrates the essentials of the three-dimensional approach and
the basic questions that are required to be answered before you start designing
or documenting about designing, which will in turn help to understand the real
requirements for a specific design:
Phase
Description
Key components
Product
Results of what?
People
Results of who?
Process
Results of how?
Before we start to apply the PPP framework on VMware vSphere, we will discuss the
list of challenges and encounters faced on the virtual infrastructure.
[7]
High utilization: An army of VMs can also throw workflows off-balance due
to the complexities they can bring to provisioning and operational tasks.
Security: More than six out of ten IT professionals believe that data
protection is a top technological challenge.
Sizing the database and mailbox: Proper sizing of databases and mailboxes
is really critical to the organization's communication systems and for
applications.
[8]
Chapter 1
What objective does the customer hope to reach with the help of a VMware
vSphere deployment?
What are the existing problems, business or technical, that the customer is
trying to resolve?
Requirement gathering and documenting are the driving factors for virtualization; the
objectives, problems and needs of the customer should be the first task to undertake
before we can start designing. This is the beginning and translates into what we have
called the design factorsrequirements, constraints, risks, and assumptions. They
are crucial to the success of any vSphere design. In the next section, you will learn the
ways in which we can overcome the challenges and enforcements.
[9]
Requirements
It's really difficult to build a virtual infrastructure if you don't know the target or
requirements. Elicitation is the step where the requirements are first gathered from
the customer. Many practices are available for gathering requirements. Each has a
certain value in certain situations; you need manifold techniques to gain a complete
picture from various sets of clients and stakeholders. Here's a look at some of the
approaches you can take:
Finally, at the end of requirement gathering, conduct workshops with all these
reviews, lastly look for room for improvement. If all stakeholders in a session
try to appraise the value and cost for each demands, at last artifacts becomes
much more effective and cost-valuable. Hence, schedule periodic meetings to
review and agree requirements:
Analysis
Research and collect data facts as far as possible before you start planning for design;
since the analysis or research must be lightweight, it must also be less scrupulous.
Best results come from considering user insights from the very start as input and
giving the result to the design process.
[ 10 ]
Chapter 1
In the early stages of the process, the discovery stage will help you identify new
breaks and will describe the design in brief. VMware vExperts advise re-iterative
ESXi and VM configuration testing before it is rolled out to the Proof of Concept
(POC). In turn, this will help to inform decisions from the design phase to the final
implementation. A usability problem can be corrected more swiftly and with less
cost if it is identified at the design stage itself rather than at the support stage.
We will use the 4D-stage analysis process to display how we can fit in all the stages:
The preceding 4D stage analysis, as explained in the following table, aligns to the
method that is required to be adopted in each stage:
Phase
Description
Methods
Discover
Cognitive Map
Define
Develop
Prototyping
Deliver
With this approach, we will be able to gain an insight into requirements and the
output of this phase is an input to the planning phase. As the next step, let's learn
about planning and its implications to the design phase.
[ 11 ]
Planning
The design considerations we made in final steps may make us to change the
decisions which we made in the earlier steps, due to various thought clashes. Every
alert challenges the potential requirement gathering, planning, and design, which
will be a great challenge throughout the project planning.
We will come to design phase which meets best of our requirements, only after
iterating through the stages as many times as required to include all of the thoughts
within the manuscript. The following diagram illustrates the key technical metrics
that are required to be a part of your planning technical artifacts:
Each phase of planning and its mission is listed in the following table:
Phases of planning
Key objectives
Chapter 1
Phases of planning
Key objectives
If you're looking for a tool that can do this, then the best tool to use (and one that
VMware recommends) is VMware Capacity Planner. It is the right thing to start with.
VMware Capacity Planner is a capacity planning product that gathers
comprehensive resource consumption data in assorted IT environments and
associates it to industry standard position data to provide analysis and decision
support modeling. The following are the key benefits of the product:
The output of data gathering from the requirements, analysis, and planning phases is
a design phase input. This is really a key factor for your success. In the next section,
let's explore the key requirements for designing in general and then we will map that
to the streamlined process and principles.
Designing
Design essentials are what this book is all about. Let's get started by looking at the
fundamentals. Any design, irrespective of the platform, should be enabled with
factors such as balance, proximity, alignment, repetition, contrast, and space. These
factors, in turn, should focus on the following:
Identifying the best practices that can fit the customer's needs.
The preceding components will help you to attain the following deliverables
that will help you to build the right solution document for small, medium, and
enterprise businesses:
[ 14 ]
Chapter 1
[ 15 ]
Readiness
As we make decisions that will form a VMware vSphere design, one value to be
considered most is readiness.
Readiness is all about integrating different components together: uptime and system
reliability, downtime, redundancy, and resiliency. In a nutshell, a mechanism that
avoids crashes and maintains performance.
Values should have a close relationship with each other. With regard to the PPP
framework of design, the value of readiness typically has the greatest effect on
product decisions.
In some cases, the functional requirements explicitly call for readiness; for example,
a functional and performance requirement may demand that the vSphere design
must provide 99 percent readiness to meet any demand. In this case, readiness is
explicitly noted by the requirement gathering and therefore must be incorporated
into the design.
In some other cases, the functional requirements may not explicitly state readiness
demands at 99 percent. In these cases, the designer should include an appropriate
level of availability as per the projected target state.
To accommodate both cases in the design document, always come up with an
assumption section. An assumption section can offer justification for the vSphere
designer's decisions within the broader business framework of functional performance
requirements, design consideration, and constraints. When readiness isn't explicitly
stated, the designer can provide an assumption that the infrastructure will be made as
ready as possible within the cost limitations of the statement of work.
Performance
Performance is often called for in the performance requirements, and it can affect
decisions in the product and people phases. For the product phase, it can affect all
manner of design decisions, starting from the type of the server's hardware to the
kind of network storage devices planned. With regard to the performance demand,
it's mostly seen as a Service Level Agreement (SLA) that expresses performance
metrics, such as response times per request, transactions per second in the network,
and the capacity to handle the maximum number of users.
If there is no performance requirement explicitly defined, the designer of
a vSphere infrastructure should consider performance as a key component
for its design decisions.
[ 16 ]
Chapter 1
The KA key stage is to review and agree the performance metrics and make sure
they have been documented on a low-level design, as well as ensuring that the
specified components are validated against the gathered requirements. The key
components that need to be considered and validated are as follows:
For Network adapters, the designer should consider the following; such as
Checksum Offload, TCP Segmentation Offload (TSO), the ability to handle
high-memory DMA (that is, 64-bit DMA addresses), the ability to handle
multiple Scatter Gather elements per Tx frame, Jumbo frames (JF), and Large
Receive Offload (LRO). When using VXLAN, the NICs should support
offload of encapsulated packets.
For memory, the designer should consider factors such as NUMA, paging
size, memory overhead, and memory swapping.
For storage, the designer should consider options such as vSphere Flash Read
Cache (vFRC), VMware vStorage APIs for Array Integration (VAAI), LUN
Access Methods, virtual disk modes, linked clones, and virtual disk types.
For network, the designer should consider the following such as Network
I/O Control (NetIOC), DirectPath I/O, Single Root I/O Virtualization
(SR-IOV), SplitRx Mode, and Virtual Network Interrupt Coalescing.
Manageability
The designer should consider that the management pane has a key component that
should be documented in a design document. Manageability is a key factor that
works on support and operations. As a part of the support, managing infrastructure
is most essential; hence, manageability is integrated from various factors:
Recovery
Recovery after a disaster is a key design principle that should be considered. With
traditional disaster recovery procedures, the IT team have to identify a way for
budgets to support the solution and make decisions about exactly which systems will
not be sheltered. This process made ease via virtualization and along with a arrival
of disaster recovery offerings via private and public clouds. Resiliency services and
business continuity services are now ease for administer to manage, in turn makes
huge disaster recovery protection for all applications.
To protect critical data and systems, only a few tools are available in the market today,
especially to query real-time data replication and failover. It is critical for vSphere
designers to understand that these tools are required for the appropriate infrastructure,
and it is very important for designers to know exactly which systems and data need
to be protected along with the type of protection that they are demanded, defining the
metrics become one of the main design essentials for vSphere.
To start with, the first key metric is the Recovery Point Objective (RPO). For any
data protection tool, RPO defines how much data could be lost in the recovery
procedure. In case a single disk drive in the RAID 5 array undergoes failure, then
there is no associated data loss. This is called as RPO of zero. At worst case scenario
is that the entire disk in an array undergoes failure. In this situation, if a business
demands data restoration from the last good known tape backup, the RPO time for
this would be 12 to 24 hours. Essential of defining RPO of VMs is highly demanding,
along with an integrating of tools in the design document.
The next serious metric is Recovery Time Objective (RTO). Used along with RPO,
RTO measures the time taken for the entire recovery process up until the time where
the end user can reconnect to their applications.
For example, if a production server fails and reaches a state where the server should
be reimaged using the same tape backup solution, the RTO could be hours or days.
This is will depend on time taken to repair the existing hardware or build the new
hardware, followed by installing the host, guest OS and followed by applications
deployment time . Hence, integrating this information on your design document
is essential.
Once recovery objectives such as RPO and RTO have been finalized, most
organization discovers that a traditional tape backup alone is not enough to
meet their objective of high availability for many critical applications, Therefore,
many organization will implement integration of different solutions such as data
replication and application high availability attached with guest OS virtualization.
[ 18 ]
Chapter 1
Security
Security is yet another key principle that has to be incorporated into virtualization
design. VMware has its own product for securityVMware vShield Endpoint
(VSE). VSE offers endpoint protection by command of magnitude and divests
anti-threat agent processing inside VMs that reside on ESXi and safe virtual
appliance brought by VMware partners. VSE improves amalgamation ratios,
performance, rationalizes antivirus deployment, monitoring for threats, and
make sure compliance by logging antivirus and anti-malware activities which
are accomplished in the virtual infrastructure.
Virtualization security is a key component of design. While the IT team needs
to implements its own security practices and security governance in the phase
of virtualization, the clear impact is that security can be really advantageous.
Virtualization improves security by assembly it more fluid and context-aware
solution. Which means security is more precise to manage, and cost effective to
deploy than traditional security software.
Virtualization security empowers datacenter admins with the authority of
automation on provision VMs. These VMs are fully wrapped with security policies
that follow desktops when users move across from one system to another system.
With the accurate processes and products, virtualization has the authority to make
datacenters even more secure; hence they identify the right tools to protect VMware
vSphere-based virtualizations.
Assumptions
Assumptions should be an important section in the design document; without the
assumption section, the planning phase cannot be driven forward. Factors that need
to be considered by designers are the following:
The existing infrastructure that can be used for the virtualization project
Server hardware that has to be separated between DMZ and internal servers
In scope and out of scope that has to be reviewed and agreed on between the
project sponsor and project supplier
If you design your capacity in such a way that your business is going to grow by
20 percent yearly, this growth will apply linearly to the ingesting of the available
capacity in the datacenter. This can be categorized under assumptions.
Constraints
Constraints are a key principle that should be considered on design essentials. This
section will impose obstacles and limit your design decision; it can be a technical
limitation, the actual cost of the project, or the choice of vendors. Here are some
examples of factors that limit the design:
Any vendor support needs for apps, servers, storage, and networks
are limited
If you purchased a SAN 3 years ago for a file server consolidation project, you plan
to use it for a virtualization project, and you do not find any information about those
devices, this can be categorized under the constraints section and it is really a critical
component for the project execution phase.
Risk
Risk management is a key principle that is required to be considered while designing
VMware vSphere infrastructure. Identifying the risk factor very early will help you
to identify and highlight problems hiding inside your virtualization design effort:
The major risk listed in the virtualization project is whether the project will
be delivered on time and on budget
Chapter 1
For example, you might be in a situation where the IT Management does not provide
the required approval for changing the network core to increase iSCSI traffic. This
can be characterized under risks on your design principles.
There are eight values of design as an outcome of design decisions. There are
multiple ways to fulfill the functional and performance requirements, but a
vSphere designer must evaluate each of the following against these eight values:
Does the choice positively or negatively impact the readiness of the design?
These principles provide generic guidance and direction on the best way to meet the
functional and performance requirements for a particular design.
Before we wrap up this chapter, we will look at a final section on the process of
designing VMware vSphere with its essentials.
[ 21 ]
Discovery
Discovery is the key, and first stage of your design. This element will help you to get
the following questions answered:
Before you start building a design you need to build up a blueprint of what
the context for that design is. This means lots of end user research, and a close
assessment of current infrastructure and policies, ethics, and business needs.
Requirement gathering
Let's get started with requirement gathering, which is an important part of any
IT virtualization project. Understanding what those requirements will bring, is
success to virtualization project. Shockingly it's an part that is frequently given
far too little courtesy.
Numerous designs start with the greenest list of demands, where some of via verse,
for those we need focus and avoid these bottleneck ,by producing a document of
requirements. This requirement document will acts as bible for design phase. It will,
in turn, provide the following benefits as well:
[ 22 ]
Chapter 1
Here are well-known rules that will you help you successfully complete your
requirement gathering:
Review, define and agree the scope of work for the virtualization
Technical assessment
Technical assessment is a critical step between the planning and design of a
virtualization project. It has three important modulesinfrastructure assessment,
business assessment, and operational assessment. The goal of the technical assessment
is to inventory current infrastructure and identify virtualization candidates.
The key deliverables of this assessment will be as follows:
Baseline the list of servers, that are ideal candidates for virtualization
Data collection does not have the arrangement required for the virtual server farm
design. Inventory data may encompass fault values and performance reports may
have wrong data, because the monitored server was shut down or not reachable via
the network. Moreover, performance reports extracted from servers having disparate
configurations cannot be equated unless data is standardized. Indeed, we cannot
compare processor utilization reports for servers with different types of processor.
This phase is predominantly interesting when the server landscape whose data
is being analyzed contains a number of servers. For each server, many data facts
are available, requirements will streamline the metrics that are required to be
considered; however, for the designer it is worth considering the following tools for
capacity planning and utilization measurements:
After you've gathered the information required to determine the design's functional
requirements, it's then required to assess the current infrastructure. Assessing the
infrastructure fixes a couple of gaps and, to do that, use the gap analysis as per the
customer's demand.
Design integration
Design integration is essential for the successful design of VMware vSphere.
Integration of design can happen only if factors such as data architecture, business
architecture, application architecture, and technology architecture are put together,
as illustrated in the following diagram:
[ 24 ]
Chapter 1
[ 25 ]
Summary
In this chapter, we discussed the VMware vSphere landscape, designing VMware
vSphere, and a list of challenges and encounters faced on the virtual infrastructure.
Designing accelerates the solutions that resolve real-world obstacles, values, and
procedures that need to be followed while designing VMware vSphere.
In the next chapter, you will learn about designing essentials for the management
layer, which is, designing the VMware ESXi host, cluster, and vCenter.
[ 26 ]
www.PacktPub.com
Stay Connected: