I.J.

Information Technology and Computer Science, 2012, 12, 21-30

Published Online November 2012 in MECS (http://www.mecs-press.org/)
DOI: 10.5815/ijitcs.2012.12.02

A Proposed Virtualization Technique to Enhance

IT Services
1
Dr Nashaat el-Khameesy & 2 Hossam Abdel Rahman Mohamed
1
Prof. and Head of Computers & Information systems Chair, Sadat Academy
2
Computer & Information System Dept - Sadat Academy, Computer & Information System Dept - Sadat Academy for
management Science –Maady-Cairo-Egypt
1
[email protected] & 2 [email protected] , [email protected]

Abstract— System virtualization is an antique art that approach of making one resource look like many or
will continue as long as applications need isolation and many resources look like one. For example, Virtual
performance independence. Virtualization provides Machines have long been used to make a single
many benefits and greater efficiency in CPU utilization, computer look like more than one (and Virtual Storage
greener IT with less power consumption, better was used to divide up the real storage in a computer for
management through central environmental control, each of the virtual machines). Conversely, Grid
more availability, reduced project timelines by Computing makes many independent computers work
eliminating hardware procurement, improved disaster together as a single, virtual, computer. Networked
recovery capability, more central control of the desktop, storage (SAN and NAS technologies) likewise makes
and improved outsourcing services. In this paper we many separate units of storage appear as one large pool.
will deliver new and enhanced IT services in less time And, in networking, Virtual LANs take a single Local
and with greater agility. The goal of providing rapid Area Network and create logically-isolated sub-LANs
near real-time response to customer requirements is in (usually for performance or security purposes). [2]
part being achieved by taking advantage of Virtualization provides compelling business value.
virtualization. Virtualization provides many benefits, Organizations can enhance IT service performance and
including improved physical resource utilization, scalability, achieving greater capacity (server MIPS,
improved hardware efficiency, and reduced power and storage GB, network K/M/Gbps) than is available from
cooling expenses. However, it introduces another set of a single instance of a resource. Efficiency is improved
systems (virtual machines) that have to be controlled, through consolidating the workloads of several under-
provisioned, managed, updated, patched, and retired. utilized resources (storage, servers, or networks) to
This paper is organized into the following areas- An improve resource utilization and reduce overall
introduction to virtualization and its benefits, the hardware, software, and management costs.
Challenges of Virtualization techniques, an overview of Virtualization can also be used to enhance availability
the hardware virtualization technique, an overview of and security through containing untrusted applications
the software virtualization technique. The paper also (e.g. JAVA applets) in a controlled environment or
proposes an effective and flexible distributed scheme isolating networks and servers into "zones" of trust. [3]
with three phases, opposing to its predecessors. Our Virtualization can be viewed as part of an overall trend
scheme achieves reducing the workload of a data center in enterprise IT that includes autonomic computing, a
and continual service improvement. scenario in which the IT environment will be able to
manage itself based on perceived activity, and utility
computing, in which computer processing power is seen
Index Terms— Virtualization techniques, Virtual as a utility that clients can pay for only as needed. The
Machine, Hypervisors,Virtual Centre, Server Clustering usual goal of virtualization is to centralize
administrative tasks while improving scalability and
overall hardware-resource utilization. [4]
I. Introduction
Virtualization is, at its foundation, a technique for
II. Virtualization Techniques Challenges [5],[6],[7]
hiding the physical characteristics of computing
resources from the way in which other systems, While virtualization offers a number of significant
applications, or end users interact with those resources. business benefits, it also introduces some new
This includes making a single physical resource (such management challenges that must be considered and
as a server, an operating system, an application, or planned for by companies considering a virtualization
storage device) appear to function as multiple logical strategy. The key virtualization technique challenges for
resources; or it can include making multiple physical companies adopting virtualization include:
resources (such as storage devices or servers) appear as
a single logical resource. [1] Virtualization is the

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30
22 A Proposed Virtualization Technique to Enhance IT Services

2.1 Bandwidth Implications need to have the capability to change dynamically to

respond to peaks and troughs in load characteristics.
Enterprises should make sure they have the
Management tools will also be required to automate
appropriate network bandwidth for their virtualization
physical to virtual, virtual to virtual, and virtual to
requirements. For example, instead of one server using
physical migration.
a 100Mb Ethernet cable, now 10 or even 100 virtual
servers must share the same physical pipe. While less of
a problem within the data center or for communication 2.5 Security
between virtual servers running in a single machine,
While virtualization can have many worthwhile
network bandwidth is a significant issue for application
security benefits, security also becomes more of a
streaming and remote desktop virtualization. These
management issue in a virtualized environment. There
technologies deliver quite substantial traffic to end users,
will be more systems to secure, more points of entry,
in most cases significantly higher than is required for
more holes to patch, and more interconnection points –
standard-installed desktop computing. Streaming
across virtual systems (where there is most likely no
technologies, although in many cases more efficient
router or firewall), as well as across physical systems.
than complete application delivery, also impose high
Access to the host environment becomes more critical,
bandwidth requirements.
as it will often allow access to multiple guest images
and applications. Enterprises need to secure virtual
2.2 Image Proliferation images just as well as they secure physical systems. [8],
[9].
Operating system and server virtualization can lead to
a rapid proliferation of system images, because it is so
much easier and faster to deploy a new virtual image
than to deploy a new physical server, without approval III. Virtualization Technique
or hardware procurement. This can impose very high
management and maintenance costs, and potentially Virtualization refers to the abstraction of logical
lead to significant licensing issues including higher resources away from their underlying physical
resources to improve agility and flexibility, reduce costs,
costs and compliance risks. This proliferation also leads
to significant storage issues, such as competing I/O and and thus enhance business value. Virtualization allows a
extreme fragmentation, requiring much faster and multi- set of underutilized physical infrastructure components
to be consolidated into a smaller number of better
channel disk access, and more maintenance time, effort,
and cost. Enterprises need to manage their virtual utilized devices, contributing to significant cost savings.
environment with the same level of discipline as their
physical infrastructure, using discovery tools to detect 3.1 Hardware Virtualization
and prevent new systems from being created without
following proper process. Hardware virtualization means that an application
executes on virtualized hardware as opposed to physical
hardware. ―Virtual hardware‖ seems to be a
2.3 Human Issues contradictory term because the hardware is supposed to
Enterprises should not underestimate the potential for have a physical existence. Furthermore, the terms
―virtual hardware‖ and ―hardware virtualization‖ are
human issues to affect their virtualization plans
adversely. Virtualization requires a new set of skills and used for different concepts in literature. We will discuss
methodologies, not just within IT, but often (certainly in in this section the techniques of Memory virtualization,
Power virtualization, Network virtualization and
the case of application and desktop virtualization) in the
end-user community. Perhaps most importantly, this Storage virtualization.
new technology requires new and creative thinking, not
just new training and skills. 3.1.1 Network Virtualization Technique [11]
Network virtualization techniques can be leveraged
2.4 Policy-Based Management in the networking infrastructure to achieve the same
Enterprises should look to deploy automated policy benefits obtained through server and storage
based management alongside their virtualization virtualization. Moreover the network, especially in the
data center, must also support this new dynamic
strategy. Resource management, for example, should
include automated policy-based tools for disk allocation environment where the computing and storage
and usage, I/O rates, CPU usage, memory allocation infrastructures are consolidated and virtualized, to meet
new requirements (VM mobility, for example) and
and usage, and network I/O. Management tools need to
be able to throttle resources in shared environments, to facilitate the delivery of IT services across
maintain service levels and response times appropriate heterogeneous network access technologies to multiple
end user devices.
to each virtual environment. Administrators should be
able to set maximum limits, and allocate resources Network interface virtualization - these techniques
across virtual environments proportionally. Allocations refer to the Ethernet NICs and how they can be

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30
 A Proposed Virtualization Technique to Enhance IT Services 23

partitioned or aggregated. Sometimes these features are operating systems. Applications see a contiguous
categorized as I/O Virtualization techniques. address space that is not necessarily tied to the
underlying physical memory in the system. The
Network Link virtualization - these techniques refer
operating system keeps mappings of virtual page
to how physical wires can be logically aggregated or
numbers to physical page numbers stored in the page
partitioned to increase throughput and reliability or to
tables. All modern x86 CPUs include a memory
provide traffic separation using the same physical
management unit (MMU) and a translation look-aside-
infrastructure.
buffer (TLB) to optimize virtual memory performance.
[12]
Network Node virtualization - these techniques refer
to how network devices can be logically aggregated.
To run multiple virtual machines on a single system,
Data Centre Network-wide virtualization- these another level of memory virtualization is required. In
techniques extend the virtualization domain to the other words, one has to virtualize the MMU to support
whole Data Centre Network and even to the WAN in the guest OS. The guests OS continues to control the
case there are multiple Data Centers. mapping of virtual addresses to the guest memory
physical addresses, but the guest OS cannot have direct
access to the actual machine memory. [13] The VMM is
3.1.2 Memory Virtualization Technique responsible for mapping guest physical memory to the
This involves sharing the physical system memory actual machine memory, and it uses shadow page tables
and dynamically allocating it to virtual machines. to accelerate the mappings. As depicted by the red line
Virtual machine memory virtualization is very similar in Figure 1.
to the virtual memory support provided by modern

Fig. 1: VMM uses TLB hardware to map the virtual memory directly to the machine memory

The VMM uses TLB hardware to map the virtual enablement and value-added software. There are three
memory directly to the machine memory to avoid the versions of Power VM, suited for different purposes:
two levels of translation on every access. When the
Power VM Express Edition- This edition is intended
guest OS changes the virtual memory to physical
for evaluations, pilots, proof of concepts, generally in
memory mapping, the VMM updates the shadow page
single-server projects.
tables to enable a direct lookup. MMU virtualization
creates some overhead for all virtualization approaches, Power VM Standard Edition- This edition is intended
but this is the area where second generation hardware for production deployments, and server consolidation.
assisted virtualization will offer efficiency gains. [14].
Power VM Enterprise Edition- the Enterprise Edition
is suitable for large server deployments such as multi-
3.1.3 Power Virtualization Technique server deployments and cloud infrastructure.
The Power VM platform is the family of technologies,
capabilities, and offerings that deliver leading 3.1.4 Storage Virtualization Technique
virtualization on the Power Systems .It is the new
Storage virtualization refers to the process of
umbrella branding term for Power Systems
Virtualization (Logical Partitioning, Micro- completely abstracting logical storage from physical
Partitioning™, Power Hypervisor, Virtual I/O Server, storage. The physical storage resources are aggregated
into storage pools, from which the logical storage is
Live Partition Mobility, Workload Partitions, and so on).
[15], [16]
as with Advanced Power Virtualization in the created. It presents to the user a logical space for data
past, Power VM is a combination of hardware storage and transparently handles the process of
mapping it to the actual physical location. This is
currently implemented inside each modern disk array,

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30
24 A Proposed Virtualization Technique to Enhance IT Services

using the vendor's proprietary solution. However, the PC management, including application and operating
goal is to virtualize multiple disk arrays, made by system (OS) migrations[18].
different vendors, scattered over the network, into a
Application Virtualization Advantages:
single monolithic storage device, which can be
managed uniformly. Minimize application conflicts and regression testing-
By reducing the requirement to install applications on
Currently, storage is virtualized by partitioning the
desktops or terminal servers, and shielding the OS and
SAN into separate partitions called Logical Unit
applications from changes created when applications
Numbers (LUNs). Each LUN can only be connected to
are installed, Application Virtualization minimizes the
one server at a time. [17] This SAN is connected to the
problems that hinder application deployments. This also
server through a Fibre Channel connection switch.
reduces lengthy application to application regression
The advantages of SAN virtualization testing. Now, applications that would traditionally
are:(Centralized stored data, Ease of backup of stored conflict with each other can easily co-exist on a single
data, Ability to remap a LUN to a different server in desktop or Terminal Services session.
case of server failure, Storage appears local to servers
Simplify OS migrations and patching- Turn time-
and to users, Improves utilization and reduces storage
consuming, tedious migration and patching projects into
growth, Reduces power and cooling requirements)
largely automated, conflict-free processes.
Build business continuity for applications- Replicate
3.2 Software Virtualization
our virtualized applications like any other enterprise
Software virtualization is the virtualization of data to maintain a rapid failover plan for your
applications or computer programs. The concept is applications, significantly cutting end-user downtime. If
similar to hardware virtualization where physical you configure Application Virtualization user profiles
machines are simulated as virtual machines. Software to persist on the network, all user-specific virtual
virtualization involves creating a virtual layer or virtual application preferences can also easily be replicated to a
hard drive space where applications can be installed. backup site.
From this virtual space, applications can then run as
though they have been installed onto host OS. We will
3.2.2 Server Virtualization Technique
discuss in this section the techniques Application
virtualization, Operating System virtualization, Desktop Server virtualization is a method of abstracting the
virtualization and Server virtualization. operating system from the hardware platform. This
allows multiple operating systems or multiple instances
of the same operating system to coexist on one or more
3.2.1 Application Virtualization Technique CPUs. A hypervisor or virtual machine monitor (VMM)
is inserted between the operating system and the
An Application Virtualization transforms
hardware to achieve this separation. These operating
applications into virtualized, network-available services
systems are called ―guests‖ or ―guest OSs.‖ The
resulting in dynamic delivery of software that is not
hypervisor provides hardware emulation to the guest
installed, minimizes application conflicts, and reduces
operating systems. It also manages allocation of
costly application compatibility testing. Users and their
hardware resources between operating systems. [19], [20],
application environments are no longer machine- [21]
specific, and the machines themselves are no longer
user-specific, enabling IT to be flexible and responsive Currently there are three common types of
to business needs, and significantly reducing the cost of hypervisors: type 1, type 2, As we see Figure2

Fig. 2: The types of hypervisors

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30
 A Proposed Virtualization Technique to Enhance IT Services 25

Type 1 - Virtualization code that runs directly on the Risk ‑ free server workload rollout, Maximizes boot
system hardware that creates fully emulated instances of performance, Minimizes network load, Provisions
the hardware on which it is executed. Also known as multiple servers from a single virtual disk image,
―full,‖ ―native,‖ or ―bare metal.‖ Minimizes storage and image management
Type 2 - Virtualization code that runs as an requirements, Ensures server build consistency, Allows
application within a traditional operating system switching of server workloads in the time it takes to
environment that creates fully emulated instances of the reboot, Maximizes flexibility by enabling real-time
hardware made available to it by the traditional changes to data centre workloads, Reduces the number
operating system on which it is executed. These are also of servers needed to support disaster recovery &
known as ―hosted‖ hypervisors. business continuity requirements.

Containers - Virtualization code that runs as an

application within a traditional operating system that 4.2 Improved Total Cost of Ownership
creates encapsulated, isolated virtual instances that are By decreasing management costs and increasing asset
pointers to the underlying host operating system on utilization, we can experience a rapid return on
which it is executed. This is also known as ―operating investment (ROI) with virtualization. In addition, by
system virtualization.‖ virtualizing resources and making them easier to
migrate or fail over to other physical devices or
3.2.3 Desktop Virtualization Technique [22] locations, you can enhance system availability and help
lower the cost and complexity of disaster-recovery
Desktop Virtualization enhances deployment and solutions.
management of Virtual PC images on a Windows
Desktop while also providing a seamless user
experience on a Virtual PC environment independent of 4.3 Increased Flexibility
the local desktop configuration and operating system. Virtualization supports the pooling of resources that
can be managed centrally through an enterprise hub to
better support changing business requirements
Desktop Virtualization Advantages dynamically.
1) Easily deploy managed Virtual PCs to any
desktop, even in less controlled environments 4.4 Enabled Access through Shared Infrastructure
such as subsidiaries, branch offices and offshore
operations. By providing a resilient foundation and shared
infrastructure that enables better access to infrastructure
2) Drive business continuity by rapid reconstitution and information in support of business applications and
of corporate Desktops. service-oriented architectures (SOA).
3) Accelerate migration to the Windows by
minimizing application- to-OS compatibility 4.5 Energy Efficiency
issues.
We can help control infrastructure sprawl through the
4) Speed user adoption of desktop virtualization by deployment of virtual servers and storage that run
making Virtual PCs ―invisible‖ to users. securely across a shared hardware environment.
Virtualization not only helps with server consolidation,
5) Reduce IT investment in desktop image
but also server containment when deploying new
management by delivering virtual images
systems. Consolidating to a virtual infrastructure can
independent of hardware or the local desktop
enable you to increase server utilization rates from 5%
configuration.
to 15% to over 70%, thus helping improve ROI. In
addition, a simplified infrastructure can help lower
manage costs with a common management platform and
IV. Design Goals
tooling.
To Decrease the workload for data center and ensure
continual service improvement, we aim to design
4.6 Rapid Application Deployment
efficient mechanisms for reducing the workload of a
data center and business Continuity verification and We can use the virtualization can help enable rapid
achieve the following goals: infrastructure provisioning (ex. Minutes compared to
days). It can help developers to speed application test
and deployment, enhance collaboration and improve
4.1 On-Demand Data Center Workload Streaming
access to the infrastructure. The ease and flexibility of
By Delivers the OS, applications, and server creating and reconfiguring guest operating systems
configuration information in a real ‑ time stream, means that development and test environments can
maximizing performance and minimizing network load, realize significant benefits from virtualization.

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30
26 A Proposed Virtualization Technique to Enhance IT Services

4.7 Improved Disaster Recovery implemented in phases allowing for capital expense
spending flexibility. The purpose of virtualization is to
Virtualization can help IT Dept secure and isolate
increase efficiency and reduce costs, with each phase
application workloads and data within virtual servers
addressing different aspects of specific resources. Phase
and storage devices for easier replication and restoration.
I should include hardware consolidation and
This added resiliency can provide IT managers with
virtualization of the web and application tiers along
greater flexibility to maintain a highly available
with the commoditization of the database tier. The
infrastructure while performing planned maintenance,
specific focus is on reducing hardware maintenance
and to configure low-cost disaster-recovery solutions.
costs, (End of Support Life) hardware and reducing
power consumption. The main objective is to increase
4.8 Managing a Virtualized Infrastructure efficient use of CPU capacity and to create system
flexibility and resiliency. Phase II should include
We can address configuration, deployment, implementation of desktop virtualization with a primary
monitoring, workload management and additional focus on change management issues and reducing the
management function in a consistent and common way
impact of human error. Further, an emphasis on training,
across their infrastructures. This can help simplify organizational restructuring and group consolidation to
problem determination, increase productivity and lower support operational technologies implemented in Phase
management costs.
1 should be incorporated. Phase III should include full
data tier virtualization and disaster recovery focusing on
full multi-site recovery in a cost effective manner.
V. Proposed Solutions
The Virtualization can reduce system and workload
of a data center. Further, virtualization can be

Table: 1 Proposed Solution Phases

Target Resource and Objective Focus

Reduction in hardware and maintenance costs
Network Infrastructure Resolve issues
Phase I
Servers/Web/Application virtualization Improve system flexibility, resiliency and efficiency in power consumption and CPU
capacity

Human Capital Improve Efficiency in Time Management

Phase II Desktop virtualization and Human Train/Retrain labor in new technology to reduce knowledge gap
Capital Investment Organizational restructuring to reduce labor costs

Fully capitalize on virtualization through integrating sites.

Reduce hardware and maintenance
Multi-site Infrastructure Costs at disaster recovery site
Phase III
Disaster Recovery Eliminate EOSL issues at Disaster
Recovery site
Enhance disaster recovery by reducing down time and implementing full recovery

5.1 Planning Phase I: At each site we separated the virtual machine and
Management Infrastructure onto two separate
The solution we offered in this paper was effectual to
subnets/VLANS and provided a third for the router to
choose the more complex option of changing the IP
use for inter site traffic.
addresses of the servers. The alternative was to assume
VLAN was stretched or moved during failover which
would be simple to document but we felt more value
5.1.1 Network infrastructure:
would be derived from recoding the IP address of each
server. Modern applications are almost always dependent on
networking to some degree or another and so physically
From a high level perspective we selected a discrete
moving a service from one site to another represents a
IP address range in each site for both the VMware
networking challenge whether the services are hosted in
Infrastructure management infrastructure and the guest
virtual machines or physical machines alike. Network
operating systems. I.e. in a failover we change all of the
addressing is not a general location independent mainly
protected virtual machines IP address.
due to poor application use of network addressing
From a VMware Infrastructure perspective all of the technologies that have been available for many years.
backing infrastructure is set up as part of the build out However we have to deal with applications and
and remains static during the failover events so this configurations which depend on hard coded IP
aspect is very straightforward. addresses.

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30
 A Proposed Virtualization Technique to Enhance IT Services 27

The solution we offered in this paper was effectual to processors were barely at 50 percent capacity
choose the more complex option of changing the IP and port utilization was at 40 percent. The
addresses of the servers. The alternative was to assume expansion potential of the system allows the
VLAN was stretched or moved during failover which organization a cost effective way to scale up, if
would be simple to document but we felt more value (or when) the need arises.
would be derived from recoding the IP address of each
server.  Database administrators are now relieved of the
time consuming task of configuring each
 From a high level perspective we selected a hardware element in the stack independently by
discrete IP address range in each site for both virtue of the Cisco UCS Manager. Further, the
the VMware Infrastructure management UCS Manager consistently and accurately
infrastructure and the guest operating systems. configures servers through service-based
I.e. in a failover we change all of the protected profiles thereby eliminating human error almost
virtual machines IP address. completely and simplifying the scaling process.
Finally, the ―wire once‖ configuration simplifies
 From a VMware Infrastructure perspective all of cabling in the rack as network cabling is done
the backing infrastructure is set up as part of the once with changes implemented through
build out and remains static during the failover software, reducing the potential error rate in the
events so this aspect is very straightforward. cabling process. Reduction in human error alone
 At each site we separated the virtual machine allows for better-managed and more effective
and Management Infrastructure onto two work time resulting in higher labor efficiency. [25]
separate subnets/VLANS and provided a third
for the router to use for inter site traffic.
5.2 Planning Phase II:
Phase two of optimization can be a springboard to
5.1.2 Infrastructure Services catapult newer employees into new technology and
Demand for virtualization of networks, servers, specialized positions all but eliminating seniority and
storage, desktops and applications continue to increase. property rights to specific positions. The notion being
Cisco is assuming the virtualization of DC services is that new technology creates new positions with
now the default building block of the data center, and specialized skills and requirements for the job. The
has organized its portfolio to hasten this transition. motivation for current employees will be to embrace
training, technology and apply for positions based on
 To ease information overload a cohesive knowledge and merit (not politics) or risk lay-off.
optimization system and process was chosen.
For this proposed study, using Citrix Xen Technically, this is the least difficult phase as the
Application and Oracle RAC Clusters on the focus is on desktop virtualization. However, this can be
Cisco Unified Computing System with Citrix the most difficult phase for managing. Virtualization
Xen Server VMware enterprise software works reduces desktop replacement costs, power consumption
to address two goals. First, this system eases the and costs related to desktop support. However, desktop
real time issue of a knowledge base shift virtualization equates to labor reduction and salary
through a cohesive and simplified deployment savings.
[23]
. Second, high performance is cost effective,
reducing budget concerns.
5.2.1 Desktop Virtualization
 Efficiency and performance through the Cisco
UCS is also realized through the Cisco The second half of phase two is the actual
partnership with Oracle RAC. Oracle has a implementation of desktop virtualization. The end result
―mission critical‖ focus in their database and being reduced complexity and risk from remote access,
application technologies [24] which are highly reduced power consumption, and reduction in endpoint
relevant to large corporations as well as the risk. A reduction in effort and risk from patching and
university. The Cisco UCS memory capability reducing sensitive data protection efforts and risks
coupled with a scalable storage system opens the should also be evident. This is realized through
performance floodgates to Oracle database leveraging knowledge from the consolidation and
environments allowing for unheard of work and virtualization put in place during phase I. With desktop
performance loads. Previous workload virtualization, power consumption is assumed to drop
performance testing ― included a realistic mix of 75% based on data and test results taken from phase I.
OLTP and DSS workloads, which generated a Reduced effort and risk as well as increased host
sustained load on the eight-node Oracle RAC security is assumed since there is no longer a
configuration for a period of 72 hours… far requirement for the end user to interact with their local
exceeding the demands of typical database host to assure patching. Further, due to a reduction in
deployments‖ . The results showed that the data footprint, there is a reasonable assumption of

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30
28 A Proposed Virtualization Technique to Enhance IT Services

increased data security, as the data no longer resides at sites are setup to have an active workload required to
the end point desktop or laptop. [26] run the day-to-day activities present at each site.
Additionally, at Site 1 there is a group of services that
must be capable of being failed over to Site 2 and run
5.3 Planning Phase III: there successfully. These are known as the protected
services. The design will also accommodate work loads
In the best of all possible worlds, organizations from Site 2 hosted on Site 1 although this was not
would implement backup and recovery processes for all implemented at this stage.
server workloads regardless of their perceived criticality.
Moreover, these recovery processes would be fast, have As shown in figure 3, the logical view of the
minimal impact on production operations and would be configuration we have used in this paper. Site 1 holds
recoverable with a high level of data integrity. In reality, virtual machines from an imaginary HR and Finance
disaster recovery needs must always be weighed against department and Site 2 hosts a number of Developer and
the fiscal need for cost-effectiveness. [27] other production services. Virtual machines can be
executed in either site with protection for the machines
To meet a number of the desired design criteria, we
in Site 1.
will use a two-site approach with an active Virtual
Centre instance at each location. These are referred to
as Site 1 and Site 2 throughout the documentation. Both

Figure 3 Site 1 holds virtual machines and Site 2 hosts a number of Developer and other production services

Virtual Centre Management Server Architecture data center elements, such as virtual machines,
hosts and clusters
Virtual Centre Management Server provides a
convenient centralized management cockpit for the data • Alarms and Event Management – Tracks and
centre. It aggregates physical resources from multiple warns users on potential resource over-utilization
Server installations and presents a central collection of or event conditions
simple and flexible resources for the system
• Task Scheduler – Schedules actions such as
administrator when provisioning virtual machines in the
VMotion to happen at a given time.
virtual environment.
Core Services are basic management services for a
virtual data center. They include services such as: VI. Reliability of Virtualization and Full Tolerance
• VM Provisioning – Guides and automates the Reliability - Virtualization is very reliable.
provisioning of virtual machines Virtualization can help us eliminate planned downtime,
ensure data protection, deliver high availability and be
• Host and VM Configuration – Allows the
better prepared for disaster recovery. The Fault-
configuration of hosts and virtual machines
Tolerance (FT) feature of VMWare, helps protect
• Resource and Virtual Machine Inventory against physical hardware failures, including network
Management – Organizes virtual machines and card failures, storage path failures and other hardware
resources in the virtual environment and failures.
facilities their management
Full tolerance- During server consolidation efforts,
• Statistics and Logging – Logs and reports on the one of the key concerns often is the impact failure of a
performance and resource utilization statistics of virtualization host might have to data centre operations.

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30
 A Proposed Virtualization Technique to Enhance IT Services 29

High availability ensures critical VMs are automatically VMware Infrastructure is the only offering that
restarted on another physical host should the original delivers a complete solution rather than
host running the VM unexpectedly fail. This can reduce piecemeal approach for customers to build a
the amount of downtime for the workload, as well as data center in a virtual environment.
eliminate the need for administrative intervention. [28]
XenServer has high-availability features that offer
granular policies governing the behavior of specific References
VMs after a host failure. Additionally, both Citrix and
[1] Susanta Nanda Tzi-cker Chiueh and Stony Brook,
VMware offer options for fault tolerance, a feature that
maintains mirrored instances of VMs running on A Survey on Virtualization Technologies. RPE
separate hosts. In the event of a host failure, the Report, pages 1-42, 2005
mirrored instance can maintain continuity of the [2] Amit singh,, ―An introduction to
workload. Citrix has partnered with several third-party virtualization‖,http://www.kernelthread.com/public
vendors, such as Marathon Technologies and Stratus, to ations/virtualization, 2004.
create more robust fault tolerance options for XenServer.
[3] Mueen Uddin,Azizah Abdul
Rahman,Virtualization Implementation Model for
Cost Effective & Efficient Data Centers, (IJACSA)
VII. Conclusions
International Journal of Advanced Computer
This paper proposes a Virtualization Techniques Science and Applications, Vol. 2, No.1, January
Model for reducing the workload of a data center and 2011
reducing complexity and costs while improving system
[4] T. Daim, J. Justice, M. Krampits, M. Letts, G.
capacity, flexibility, and efficiency. In addition, the
proposed model improves service Continuity. [29], [30], [31], Subramanian, M. Thirumalai, ―Data center metrics
[32] An energy efficiency model for information
technologu managers‖, Management of
We have proposed an effective and flexible Environmental Quality, Vol.20 No.6, 2009.
virtualization solutions Infrastructure that is architected
[5] Chalapathi Valupula,R. Lakshman Naik,Sunitha
from the ground up to enable organizations to bring the
benefits of virtualization to the entire enterprise Muppa, Scheduling in Virtual Infrastructure for
platform, including server, storage and networking Highthroughput Computing ,International Journal
of Computer Science, Engineering and
systems. The virtualization solutions Infrastructure not
only enables organizations to build an entire virtual data Applications (IJCSEA) Vol.2, No.3, June 2012
centre that matches the capabilities available in the [6] Virtualization: State of the Art Version 1.0, April 3,
physical world, but it goes one step further by offering 2008 Copyright © 2008 SCOPE Alliance.
the flexibility and control that is physically cost-
prohibitive or impossible. [7] VMWare. Transparent Previrtualization info page.
http://www.vmware.com/interfaces/paravirtualizati
 We proposed an Implemented plan which is on.html.
easy to manage and scalable for future growth,
[8] Haletky, Edward. (2009). VMware vSphere™ and
resulting in reduced administrative burden.
Virtual Infrastructure Security: Securing the
 Our proposed An easier, faster and consistent Virtual Environment. Prentice Hall. Retrieved June
centralized server deployment and protection of 12, 2010, from http://www.safaribooksonline.com
Virtual Machines
[9] Flavio Lombardia, Roberto Di Pietrob, Secure
 We have discussed how System Virtualization virtualization for cloud computing Journal of
Networking can help resolve network Network and Computer Applications Volume 34,
configuration challenges introduced by server Issue 4, July 2011, Pages 1113–1122
virtualization. [10] Gai, Silvano; Salli, Tommi and Anderson, Roger
 We have discussed management capabilities for (2010). Cisco Unified Computing System (UCS)
virtualization Techniques (Data Center): A Complete Reference Guide to the
Cisco Data Center Virtualization Server
 We have discussed how virtualization can help Architecture. Indianapolis, IN: Pearson Education,
automate network configuration in virtualized Cisco Press. Retrieved June 7, 2010, from
environments, and how it enhances network http://www.safaribooksonline.com
control by making the network aware of virtual
machines. [11] P. Viana. Exploring Memory Hierarchy with
ArchC. Symposium on Computer Architecture and
 With a whole suite of complementary High Performance Computing, pages 2–9, 2003.
virtualization and management services such as
virtual VMware VMotion, VMware DRS, [12] C. Arajo. Platform Designer: An Approach for
VMware HA and VMware Consolidated Backup, Modeling Multiprocessor Platforms based on

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30
30 A Proposed Virtualization Technique to Enhance IT Services

SystemC. Journal of Design Automation for [24] Cisco. (2010). Deploying Oracle Real Application
Embedded Systems, 10(4):253–283, Springer 2003. Clusters on the Cisco Unified Computing System
with EMC CLARiiON Storage. White Paper.
[13] G. Neiger. Intel Virtualization Technology:
Cisco.com. Retrieved September 1, 2010,
Hardware Support for Efficient Processor
http://www.cisco.com/en/US/prod/collateral/ps102
Virtualization. Intel Computer Journal, 10:166–
65/ps10280/white_paper_c11- 562881.pdf.
178, August 2006.
[25] MENON, A., COX, A. L., AND ZWAENEPOEL,
[14] K. Adams and O. Agesen, ―A comparison of
W. Optimizing network virtualization in Xen. In
software and hardware techniques for x86
USENIX Annual Technical Conference (June
virtualization‖ in ASPLOS-XII: Proceedings of the
2006).
12th international conference on Architectural
support for programming languages and operating [26] LEVASSEUR, J., UHLIG, V., CHAPMAN, M.,
systems (San Jose, California, USA, 2006), ACM, CHUBB, P., LESLIE, B., AND HEISER, G. Pre-
pp. 2-13. virtualization: Slashing the cost of virtualization.
Technical Report 2005-30, Fakult¨at f¨ur
[15] NEIGER, G., SANTONI, A., LEUNG, F.,
Informatik, Universit¨at Karlsruhe (TH), Nov.
RODGERS, D., AND UHLIG, R. Intel
2005.
virtualization technology: Hardware support for
efficient processor virtualization. Intel Technology [27] VMware® Fault Tolerance Recommendations and
Journal 10, 3 (2006). Considerations on VMware vSphere™ 4
http://fault_tolerance_recommendations_considera
[16] Adrian De Luca, Mandar Bhide, Storage
tions_on_vmw_vsphere4.pdf
Virtualization For Dummies, Wiley Publishing, Inc.
111 River Street Hoboken, NJ 07030-5774 [28] http://www.ibm.com/support/techdocs/atsmastr.nsf
Copyright © 2010 by Wiley Publishing, Inc., /WebIndex/WP101520
Indianapolis, Indiana
[29] http://www.ibm.com/support/techdocs/atsmastr.nsf
[17] He,Yu,Hu,Yuhong&Wang,Yibo. (2007). /WebIndex/WP101597
Application of virtualization technology in data
[30] Enabling a Dynamic Datacenter with Microsoft
center of campus network. Journal of University of
Virtualization , Published: June 2008,
Electronic Science and Technology of China.
http://download.microsoft.com/download/a/7/a/a7a
2007,36(6):1461-1464..
d1735-40d7-4508-9462-
[18] James E. Smith and Ravi Nair. The Architecture of bcb79577b17b/virtualization_whitepaper_v2_1.pdf
Virtual Machines. IEEE Computer, 38(5):32-38,
[31] Citrix- Advanced virtualization management for
2005.
Microsoft virtualization,
[19] Y. Ha, S. Vernalde, P. Schaumont, M. Engels, R. http://www.networkworld.com/chatterbox/citrix/di
Lauwereins, and H. De Man, ―Building a Virtual mension3/Citrix_Essentials_for_Hyper-
Framework for Networked Reconfigurable V_Product_Overview_March_2009.pdf
Hardware and Software Objects,‖ Journal of
Supercomputing, vol. 21, no. 2, pp. 131–144,
February 2002. Dr Nashaat el-Khameesy: Prof. and
Head of Computers & Information
[20] Wang,Jianhong. (2008). The analysis on computer
systems Chair, Sadat Academy
virtualization of linux. Journal of Hubei Normal
University (Natural Science).2008, 28(1): 62-64.
[21] A. Whitaker, M. Shaw, and S. D. Gribble. Denali:
Lightweight Virtual Machines for Distributed and
Networked Applications. Technical Report 02-02-
Hossam Abdel Rahman Mohamed:
01, University of Washington, 2002.
Post-graduate student for degree of
[22] K. A. Fraser, S. M. Hand, T. L. Harris, I. M. Leslie, Master of computer science &
and I. A. Pratt. The Xenoserver computing information system - Sadat Academy -
infrastructure. Technical Report UCAM-CL-TR- Dept Computer and Information
552, University of Cambridge, Computer System,Maady , Cairo , Egypt - He is
Laboratory, Jan. 2003. currently position Data Center Operation
Team leader at Trans IT Company - Ministry of
[23] MENON, A., SANTOS, J. R., TURNER, Y., Transportation
JANAKIRAMAN, G., AND ZWAENEPOEL, W.
Diagnosing performance overheads in the Xen
virtual machine environment. In First
ACM/USENIX Conference on Virtual Execution
Environments (VEE’05) (June 2005).

Copyright © 2012 MECS I.J. Information Technology and Computer Science, 2012, 12, 21-30