OE-CLOUD COMPUTING

MODULE_1
VIRTUALIZATION AND HYPERVISORS

Describing virtualization

• Sixties and seventies -Mainframe processing.

• Eighties and nineties -Personal computers, the digitization of the physical desktop,
and client/server technology.
• last and current centuries -Internet
• Now we in the midst of another of those model-changing trends: virtualization.

• Virtualization in computing often refers to the abstraction of some physical

component into a logical(Virtual Machine).
• Over the last fifty years, certain key trends created fundamental changes in how
computing services are provided.
Example
• Virtual LANs (local area networks), or VLANs, provide greater network performance
and improved manageability by being separated from the physical hardware.
• Similarly, storage area networks (SANs) provide greater flexibility, improved
availability, and more efficient use of storage resources by abstracting the physical
devices into logical objects that can be quickly and easily manipulated.
Our focus is on virtualization of entire computers.
• Virtual machine (VM) can virtualize all of the hardware resources, including
processors, memory, storage, and network connectivity.
• A virtual machine monitor (VMM), which today is commonly called a hypervisor, is
the software that provides the environment in which the VMs operate.

Virtualization

• Virtualization is the ability to run multiple operating systems on a single physical

system and share the underlying hardware resources.
• The machine on which the virtual machine is created is known as Host machine and
virtual machine is referred as a guest machine.
• This machine is managed by a software or firmware, which is known as hypervisor.
1
Page

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD

OE-CLOUD COMPUTING

Fundamental idea
Abstract hardware of a single computer into several different execution
environments
• Similar to layered approach
• Layer creates virtual system (virtual machine, or VM) on which applications can run
Several components
• Host– underlying hardware system
• Virtual machine monitor / manager(VMM) or hypervisor– creates and runs virtual
machines by providing interface that is identical to the host
• Guest– Virtual machine ( Usually an operating system)
• Single physical machine can run multiple operating systems concurrently, each in its
own virtual machine.

Describing a Hypervisor
• The hypervisor is a layer of software that resides below the virtual machines and
above the hardware which provides an environment for programs that are identical to
original machine with minor decreases in execution speed together with complete
control over resource allocation.
• . The hypervisor manages the interactions between each virtual machine and the
hardware that the guests all share.
 Initially, virtual machine monitors were used for the development and
debugging of operating systems because they provided a sandbox for
programmers to test rapidly and repeatedly, without using all of the resources
of the hardware.
 They added the ability to run multiple environments concurrently, carving the
hardware resources into virtual servers that could each run its own operating
system.
•A , also called as a virtual machine manager/monitor (VMM), or
virtualization manager, is a program that allows multiple operating systems to share a
single hardware host.
• Each guest operating system appears to have the host's processor, memory, and other
resources all to itself.
• Hypervisor controls the host processor and resources, by allocating what is needed
to each operating system in turn and making sure that the guest operating systems
(called virtual machines) cannot disrupt each other.
2
Page

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD

OE-CLOUD COMPUTING

Properties of VMM
• According to Popek and Goldberg
(Who described the requirements for a computer system to support virtualization.)
VMM needs to exhibit three properties in order to correctly satisfy their definition:
• Fidelity (reliability) the environment it creates for the VM is essentially identical to
the original (hardware) physical machine.
• Isolation or Safety the VMM must have complete control of the system resources.
• Performance There should be little or no difference in performance between the VM
and a physical equivalent.

Microsoft Windows Drives Server Growth

• Microsoft Windows was developed during the 1980s primarily as a personal computer
operating system.
• Companies moved from paper based records to running their accounting, human
resources, and many other industry-specific applications on mainframes or
minicomputers.
• These computers usually ran vendor-specific operating systems, making it difficult, if
not impossible, for companies and IT professionals to easily transfer information
among incompatible systems.
• This led to the need for standards, agreed upon methods for exchanging information.
So operating systems and programs should be able to run on many different vendors’
hardware. For example UNIX operating systems.

Explaining Moore’s Law

Data Center

• Windows was originally designed to be a single-user operating system, a single application on

a single Windows server ran fine, but often when a second program was introduced, the
requirements of each program caused various types of resource contention and even out and
out operating system failures.
• This behaviour drove many companies, application designers, developers, IT professionals,
and vendors to adopt a “one server, one application” best practice; so for every application
that was deployed, one or more servers needed to be acquired, provisioned, and managed.
• As companies depended more and more on technology to drive their business, they added
many more servers to support that need. Eventually, this expansion created data centers.
• A data center could be anything from a larger computer room, to an entire floor in a
building, which existed solely to support many servers.
• Data centers started to grow massively in 2006, One study estimated that the 16 million
servers in use in 2000 had grown to almost 30 million by 2005.
• A computer that you bought in 2000 has probably been supplanted by one you
3

purchased in the last three to five years, and if it is closer to five years, you are
Page

probably thinking about replacing that one as well.

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD
OE-CLOUD COMPUTING

• For example, digital cameras first captured images at less than one megapixel resolution and
now routinely provide more than 12 megapixel resolutions. PCs, and now smart phones,
Initially offered memory (RAM) measured in kilobytes; today the standard is giga bytes, an
increase of two orders of magnitude.

Moore’s Law

 It is called Moore’s Law, and it deals with the rate at which certain
technologies improve
 Moore’s Law applies not just to processing power (the speed and capacity of
computer chips) but to many other related technologies as well (such as
memory capacity and the megapixel count in digital cameras).

Understanding the Importance of Virtualization

• There was a wild explosion of data centers overfilled with servers; but as time passed,
in a combination of the effect of Moore’s Law and the “one server, one application”
model, those servers did less and less work.
• Popek and Goldberg’s definition, virtualization allows many operating systems to run
on the same server hardware at the same time, while keeping each virtual machine
functionally isolated from all the others. The first commercially available solution to
provide virtualization for x86 computers came from VMware in 2001.
Benefits of virtualization:
• Sharing- of resources helps cost reduction
• Isolation-virtual machines are isolated from each other’s as if they are physically
separated.
• Encapsulation-virtual machine encapsulate a complete computing environment.
• Hardware independence-virtual machine run independently underlying hardware.
• Portability-virtual machine can be migrated between different hosts.
Server consolidation and Containment:
• With virtualization, overfull data centers and underutilized servers was the ability to condense
4

multiple physical servers into one server that would run many virtual machines, allowing that
Page

physical server to run at a much higher rate of utilization.

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD

OE-CLOUD COMPUTING

• This condensing of servers is called consolidation, A measure of consolidation is called the

consolidation ratio and is calculated by counting the number of VMs on a server for example,
a server that has eight VMs running on it has a consolidation ratio of 8:1.

• As companies began to see the benefits of virtualization

 They no longer purchased new hard ware when their leases were over, or if
they owned the equipment, when their hardware maintenance licenses
expired.
 Instead, they virtualized those server workloads. This is called containment.

 Containment benefited corporations in multiple ways.

 They no longer had to refresh large amounts of hardware year after year; and
all the costs of managing and maintaining those servers—power, cooling,
etc.—were removed
 In larger data centers, where hundreds or even thousands of servers were housed,
virtualization provided a way to decommission a large portion of servers.
 This reduced
 the overall footprint of a data center
 reduced the power and cooling requirements
 Removed the necessity to add to or construct additional data centers.
 With fewer servers,
 it reduced a company’s hardware maintenance costs
 Reduced the time system administrators took to perform many other routine
tasks.
5
Page

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD

OE-CLOUD COMPUTING

Virtualization and Cloud Computing

• Apple for example, recently offered the iCloud where you can store your music,
pictures, books, and other digital possessions and then accesses them from anywhere.
Other companies, such as Microsoft, Amazon, and Google are offering similar cloud-
based services.
• Virtualization and, by extension, cloud computing provide greater automation
opportunities that reduce administrative costs and increase a company’s ability to
dynamically deploy solutions.

Dynamic Data Center

• Virtualization helps us break the “one service per server” model
• Consolidate many services into fewer numbers of machines.
• Conversely, as demand for a particular service increases, we can shift more virtual
machines to run that service.
We can build a data center with fewer total resources, since resources are used as
needed instead of being dedicated to single services
• This virtual data center, deployed in the cloud, offers resources on an as-needed basis, much
like a power company provides electricity.
• This new model of computing allows companies to accelerate their deployments without
sacrificing scalability, resiliency, or availability.
.
Understanding Virtualization Software Operation

• Personal computers are changing into tablets and thin clients, but the applications that
run on PCs still need to be offered to users.
• One way to achieve this is desktop virtualization. Those applications can also be
virtualized, packaged up, and delivered to users.
• Virtualization is even being pushed down to the other mobile devices such as smart
phones

1. Virtualizing Servers

• A hypervisor is installed directly onto a server, then instantiated, or booted.

• From the virtual machine’s view, it can see and work with a number of hardware
resources. The hypervisor becomes the interface between the hardware devices on the
physical server and the virtual devices of the virtual machines.
• The hypervisor presents only some subset of the physical resources to each individual
virtual machine and handles the actual I/O from VM to physical device and back
again. Hypervisors enable enhanced availability features and create new and better
ways for provisioning and management as well.
6
Page

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD

OE-CLOUD COMPUTING

• Virtual machines contain everything that their physical counterparts do (operating

systems, applications, network connections, access to storage, and other necessary
resources) but packaged in a set of data fi les.
• This packaging makes virtual machines much more flexible and manageable .
• Virtual machines can be cloned, upgraded, and even moved from place to place,
without ever having to disrupt the user applications.

2. Virtualizing Desktops

• Desktop computing for companies is expensive and inefficient on many fronts.

• It requires staffs of people to handle software update rollouts and patching processes,
not to mention hardware support and help desk staffing.
• Virtual desktops run on servers in the datacenter; these servers are much more
powerful and reliable hardware than traditional PCs.
• The applications that users connect to are also in the data center running on servers
right next door, if you will, so all of the network traffic that previously had to go back
and forth to the data center, no longer needs to, which greatly reduces network traffic
and extends network resources.
• Virtual desktops are accessed through thin clients, or other devices, many of
which are more reliable and less expensive than PCs.

3. Virtualizing Applications

• Computer programs, or applications, can also be virtualized. Like both server and
desktop virtualization, there are a number of different solutions for this problem.
• There are two main reasons for application virtualization;
1. the first is ease of deployment.
Think about the number of programs you have on your PC. Some companies must
manage hundreds or even thousands of different applications. Every time a new
version of each of those applications is available, the company, if it decides to
upgrade to that newer version, has to push out a copy to all of its PCs. For one or a
small number of computers, this may be a relatively trivial task. But how would you
do this to a hundred PCs? Or a thousand?
• 2. The second reason has to do with how different applications interact with each
other.

Types of Virtualization

• There are two different techniques of server or machine virtualization they are hosted
approach and the bare metal approach. The techniques differ depending on the type of
hypervisor used.
7
Page

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD

OE-CLOUD COMPUTING

Bare Metal Approach & Type 1 Hypervisor

• In this approach of machine virtualization, the hypervisor is directly installed over the
physical machine.
• Since, the hypervisor is the first layer over hardware resources; hence, the technique
is referred as bare metal approach.
• Here, the VMM or the hypervisor communicates directly with system hardware.
• In this approach, the hypervisor acts as low-level virtual machine monitor and also
called as Type 1 hypervisor or Native Hypervisor.
• VMware’s ESX and ESXi Servers, Microsoft’s Hyper-V, solution Xen are some of
the examples of bare-metal hypervisors.

• Benefits: Since the bare metal hypervisor can directly access the hardware resources
in most of the cases it provides better performance in comparison to the hosted
hypervisor.
• For bigger application like enterprise data centers, bare-metal virtualization is more
suitable because usually it provides advanced features for resource and security
management. Administrators get more control over the host environment.
• Drawbacks: As any hypervisor usually have limited set of device drivers built into it,
so the bare metal hypervisors have limited hardware support and cannot run on a wide
variety of hardware platform.

Hosted Approach & Type 2 Hypervisor

• In this approach, an operating system is first installed on the physical machine to

activate it.
8

• This OS installed over the host machine is referred as host operating system.
Page

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD

OE-CLOUD COMPUTING

• The hypervisor is then installed over this host OS. This type of hypervisor is referred
to as Type 2 hypervisor or Hosted hypervisor.

• Figure represents the hosted machine virtualization technique. So, here the host OS
works as the first layer of software over the physical resources.
• Hypervisor is the second layer of software and guest operating systems run as the
third layer of software.
• Products like VMWare Workstation and Microsoft Virtual PC are the most common
examples of Hosted Approach & Type 2 Hypervisor.
• In this approach, an operating system is first installed on the physical machine to
activate it.
• This OS installed over the host machine is referred as host operating system.
• The hypervisor is then installed over this host OS. This type of hypervisor is referred
to as Type 2 hypervisor or Hosted hypervisor.

CRITERIA TYPE 1 HYPERVISOR TYPE 2 HYPERVISOR

Bare metal or Native Hosted
Definition Runs directly on the system Runs on a conventional
with VMs running on them. Operating System.
Virtualization Hardware virtualization OS Virtualization
Operation Guest OS and application Runs as an application on the
runs on the hypervisor. host Os.
Scalability Better scalability Not so much, because of its
reliance on the underlying
OS.
Setup/Installation Simple as long as you have Lot simpler setup, as you
the necessary hardware have already have an
9

support. Operating System.

Page

System Independence Has direct access to hardware Are not allowed to directly
NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD
OE-CLOUD COMPUTING

along with virtual machines access the host’s hardware

it hosts. and its resources.
Speed Faster Slower because of the
system’s dependency.
Performance Higher performance as Comparatively has reduced
there’s no middle layer performance rate as it runs
with extra overhead.
Security More secure Less secure, as any problem
in the base operating system
affects the entire system
including the protected
hypervisor.
Examples VMware ESXI VMware Workstation Player
Microsoft Hyper-V Microsoft Virtual PC
Citrix XenServer Oracle Virtual Box

Q. Illustrate a guest failure in a Type 1 hypervisor?

Type 1 hypervisors are also considered to be more secure than Type 2 hypervisors. Guest operations
are handed off, and, as such, a guest cannot affect the hypervisor on which it is supported. A virtual
machine can damage only itself, causing a single guest crash, but that event does not escape the VM
boundaries. Other guests continue processing, and the hypervisor is unaffected as well. A malicious
guest, where code is deliberately trying to interfere with the hypervisor or the other guests would be
unable to do so. Figure illustrates a guest failure in a Type 1 hypervisor.

Comparing Today’s Hypervisors

VMware ESX
 ESX was a Type 1 hypervisor
 All the virtual machine or Guest OS installed on ESXi server.
 Linux-derived Service Console.
 Used to provide an interactive environment through which users could interact with
10

hypervisor.
Page

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD

OE-CLOUD COMPUTING

Citrix Xen

 Is a software layer that runs directly on the harware below any OS.
 The Xen model has a special guest called Domain 0, also referred to as Dom0.
 This guest gets booted when the hypervisor is booted, and it has management
privileges different from the other guests. Because it has direct access to the
hardware, it handles all of the I/O for the individual guests. It also handles the
hardware device driver support.
 When additional guests make requests of the underlying hardware resources, those
requests go through the hypervisor, up to the Dom0 guest, and then to the resource.
Results from those resources reverse that trip to return to the guests.

11

Microsoft Hyper-V
Page

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD

OE-CLOUD COMPUTING

 Hyper-V is a Type 1 hypervisor because the hypervisor code lives directly on the
hardware.
 The nomenclature is slightly different, though—rather than guests, the virtualized
workloads are called partitions.

12
Page

NEETHU PONNACHAN (LECTURER IN CT) GPTC NEDUMANGAD