Chapter 2

Understanding Virtualization Technologies

 Outlines
• Virtualization
• Server virtualization
• Storage virtualization
• I/O virtualization
• Network virtualization
• Client virtualization
• Application virtualization
• Desktop virtualization
 Virtualization
• Virtualization creates a virtual version of something that is generally
used for some type of execution.
• Virtualization in computing refers to the abstraction of some physical
component into a logical object.
• The two most common types of virtualization applied in the data
center are server virtualization and storage virtualization.
Virtualization (Cont’d)

Figure: Server virtualization

 Server Virtualization
• Virtualization helps to integrate several servers into a physical server in
such a way that every server appears as a standalone server to the users.
• Allows the user to run multiple Virtual Machines (VMs), each with their
own OS, on the same physical server

Virtual Host Virtual Machine

Physical server with virtualization layer Each guest OS running on the host

• Three main types of server virtualization

• Operating system virtualization
• Hardware emulation
• Paravirtualization
 Operating System Virtualization
• OS virtualization helps to create virtualized layer of software on the top of
host operating system that resides above the hardware layer.
• Unlike other virtualization, they create an OS interfaces for applications to
run, giving the feeling of a complete OS for the applications.
• Each virtualized environment has its own file system, system libraries,
process tables and network configuration.
• Since they create a self-contained environment, they are also known as
“container”.
• Therefore, creating the software emulation of an entire OS in a physical
server is the essence of OS virtualization.
 Operating System Virtualization (Cont’d)
• Advantages
• Since OS virtualization provide least overhead among all types of virtualization
solution, they offer highest performance and highest density of virtual environment.
• Disadvantages
• They support only one operating system as base and guest OS in a single server.
• All the OS in the container should be same version and should have same patch level
of the base OS.
• If the base OS crash, all the virtual container become unavailable.
• Application
• They are great for isolating applications, web hosting and training. Popular OS
virtualized products are OpenVZ and Virtuozzo.
Operating System Virtualization (Cont’d)
Virtualization Vs. Containerization
 Hardware Emulation
• In hardware emulation, a specialized software, which is also known as
hypervisor, creates hardware emulation for OS in a single server.
• They can host different types of OS in a single server.
• The OS loaded into each virtual machine works as a standalone and
unmodified OS.
• When a VM is running, the hypervisor make changes to the part of the OS
that make system calls.
• Hypervisor changes the OS by entering a piece of code—which is known
as binary translation- to the OS when it is running.
• Binary translation takes place in four parts of the OS (memory, processor,
network, storage) that interact with the hardware.
Hardware Emulation (Cont’d)

Figure: Hardware emulation virtualization

 Types of Hypervisor
• The two classes of hypervisor are: type 1 and type 2.
• Type 1 hypervisor is also known as bare-metal implementation because they
sit directly on the top of hardware, without needing any operating system.
• Since they can directly communicate with hardware resources, they are
much faster than type 2 hypervisor.
• Advantages
• If a single virtual machine crashes, it does not affect the rest of the guest operation
system.
• Therefore, they are considered more secure than type 2.
• Since they generate less overhead, type 1 hypervisor is much faster than its
counterpart.
• Type 1 products include VMware ESX, Microsoft Hyper-V, and the many
Xen variants.
 Types of Hypervisor (Cont’d)
• Type 2 hypervisor resides on top of the operating system.
• Since they cannot directly communicate with the hardware, they are less efficient than
the type 1.
• Advantages
• OS takes care of all the hardware.
• Type 2 hypervisor can support a wide range of hardware.
• Besides, installing a type 2 is much easier than that of type 1.
• Disadvantages
• They have more points of failure since anything that affect the stability of the base operating
system can also affect the guest OS and the virtual machine.
• When the base OS needs a reboot, all the VM will also be rebooted.
• Examples of type 2 hypervisor are VMware Player, VMware Workstation, and
Microsoft Virtual Server.
 Types of Hypervisor (Cont’d)

A type 1 hypervisor A type 2 hypervisor

Figure: Two types of hypervisor

 Paravirtualization
• Paravirtualization does not emulate hardware environment in software, instead it
coordinates or multiplexes access to hardware resources in favor of virtual
machine.
• A guest OS (DomainU) is directly installed on the hypervisor (bare-metal
architecture) that does not contain device drivers for network and storage.
• Instead a privileged guest OS also known as DomainO has direct access to
hardware.
• When a guest OS (virtual machine) needs to access hardware resources it sends a
message to DomainO that access the hardware on behalf of guest OS.
• When data returns from the hardware, DomainO reads the information and passes
it back to the guest operating system that requests it.
• DomainO is a standard OS such as Linux that has been modified to communicate
with hypervisor to control access to hardware.
Paravirtualization (Cont’d)

Figure: A look at paravirtualization

 Paravirtualization (Cont’d)
• Unlike hardware emulation, which has device drivers installed in the hypervisor,
paravirtualization uses device drivers of the DomainO.
• All the guest operating systems have stub drivers that communicate with the stub drivers
in the privileged guest.
• Instead of making direct contact with the hardware, each stub driver in the guest OS
communicate with its counterpart in the DomainO.
• Advantages
• The hypervisor does not have to have its own device drivers.
• Therefore, the users of the virtual machine never have to depend on the hypervisor software provider for
driver software.
• Faster run time translation for system calls is the second benefit of this approach.
• Disadvantages
• the requirement of modifying guest operating system to execute and communicate with the hypervisor
• You must modify the kernel of the guest OS before installation.
• Since, it is not possible to modify the kernel of proprietary OS like Windows, paravirtualization software
users (such as Xen users) have to use open source OS like Linux or OpenSolaris.
• Example: Xen open source virtualization software
 Storage Virtualization
• Storage virtualization is the process of abstracting logical storage from physical storage.
• The physical storage resources (such as disk drives) are aggregated into storage pools,
from which the logical storage is created and presented to the application environment.
• Storage virtualization can be implemented within the storage arrays themselves (array-
based virtualization) or at the network level where multiple disk arrays or networked
storage systems from different vendors, scattered over the network, can be pooled into a
single monolithic storage device.
• This allows the multiple arrays to be managed uniformly as if they were a single pool.
• Virtualized storage arrays offer more flexibility, simplified management, and better
performance and capacity utilization in comparison to traditional disk arrays.
• There are two major types of shared networked storage systems: NAS and SAN systems.
 Network Attached Storage
• Network-attached storage (NAS) is a storage device that sits on the network and offers
storage to servers on the network.
• It allows multiple clients, like PC users, and servers to share files over a Local Area
Network (LAN).
• NAS uses file-based protocols such as NFS or SMB/CIFS where it is clear that the
storage is remote, and computers request a file rather than a disk block.
• Moreover, with all the files moved to a central location, it is much easier to manage them,
enabling better backup, archiving, and so on.
• One advantage of NAS is that it is IP based and simple to deploy and manage.
• Common uses of NAS include rapid file storage for rich media, document and backup
files, and e-mail.
 Storage Area Network
• A storage area network (SAN) is a storage device accessible to servers so the devices
appear as locally attached to the operating system.
• A SAN typically has its own network of storage devices that are generally not accessible
through the regular network by regular devices.
• A SAN alone doesn’t provide the “file” abstraction like NAS, only block-level operations.
• Most SANs use Fiber Channel connectivity, a network technology specially designed to
handle storage communications, or iSCSI, which is an IP-based networking standard for
linking storage devices.
• Companies move to SAN storage to centralize the management of corporate data.
• Common uses of a SAN include provisioning of transactionally accessed data that
require high-speed block-level access to the storage hard drives such as e-mail servers,
databases, and high usage file servers.
 Differences between SAN and NAS
SAN NAS
Block level data access File level data access
Fiber channel is the primary media used with Ethernet is the primary media used with NAS
SAN
SCSI is the main I/O protocol NFS/CIFS is used as the main I/O protocol in
NAS
SAN storage appears to the computer as its own NAS appears as a shared folder to the computer
storage
It can have excellent speed and performance It can sometimes worsen the performance, if the
when used with fiber channel media network is being used for other things as
well(which normally is the case)
Used primarily for higher performance block Used for long distance small read and write
level data storage operations
 I/O Virtualization
• I/O virtualization is technology that uses software to abstract upper-layer protocols from
physical connections or physical transports.
• This technique takes a single physical component and presents it to devices as multiple
components.
• In I/O virtualization, a virtual device is substituted for its physical equivalent, such as a
network interface card (NIC) or host bus adapter (HBA).
• I/O virtualization is based on a one-to-many approach.
• The path between physical server and nearby peripherals is virtualized, allowing a single
IT resource to be shared among virtual machines.
• Using I/O virtualization, an IT administrator is able to spin up a large number of VMs on
an individual server, which reduces the need for new hardware.
 Network Virtualization
• Network virtualization allows the network to be reconfigured on the fly without any need
to touch a single cable or device.
• Instead, virtualization-capable network devices are managed remotely and can be
reconfigured logically.
• This ability to perform network modification remotely and logically completes the
virtualization of the data center.
• Every type of resource — from server to storage and everything in between — is no
longer physically tied to specific pieces of hardware.
• Instead, every type of resource can be addressed logically and reconfigured without any
need to physically set hands on it.
 Client Virtualization
• A vast number of client devices are used throughout companies; in many companies,
nearly every employee has his or her own PC, whether a desktop device or laptop, plus
smartphones and notepads are now in the mix.
• Keeping all those devices current with operating system patches, application updates,
virus and spyware definitions, and so on is a virtually unending task.
• Keeping track of the devices themselves and ensuring they’re kept secure has lent impetus
to the move to client virtualization.
 Application Virtualization
• Application virtualization refers to a separation of program execution from program
display; in other words, a program like Microsoft Word executes on a server located in the
data center, but the graphical output is sent to a remote client device.
• The end-user sees the full graphical display of the program and is able to interact with it
via keyboard and mouse.
• A variant of application virtualization is one in which the application doesn’t execute on a
server in the data center, but on the client device.
 Desktop Virtualization
• Unlike application virtualization where one or more applications are displayed or
streamed from a central server, in desktop virtualization a user’s entire PC executes on a
central server, with the graphical display output to a client device.
• This form of client virtualization referred to as VDI (Virtual Desktop Infrastructure).
• It is easier to keep client systems updated with patches.
• Instead of needing to store one desktop image for each user, much of each image is
identical — the latest form uses one single image that’s cloned as required.
• This cloning cuts down enormously on the amount of needed storage and makes the
economics of desktop virtualization even more attractive.
• Desktop virtualization often uses an inexpensive client device for the end-user display
and interaction.
• These thin clients can be cheap devices with little computing power and no local disk
storage.
• This can reduce the cost per end-user device significantly.
Thank You