THE IMPACT OF SERVER VIRTUALIZATION

ON ITIL PROCESSES

Jurriaan Kamer and Harald Vranken

Open Universiteit, P.O. Box 2960, 6401 DL Heerlen, The Netherlands

Keywords: Virtualization, ITIL, IT service management, Cloud computing.

Abstract: Server virtualization influences all aspects of IT service management, and is a key enabling technology for

cloud computing. In this paper we focus on the impact of server virtualization on service delivery and

service support as described by ITIL. We identify advantages, disadvantages, and risks of server

virtualization for capacity, management, availability, costs, and security of IT services, and relate these

aspects to the ITIL processes. We validated our results using an empirical test within four different

organizations. Our main conclusion is that server virtualization does not change the ITIL processes

themselves, but it does change the way the processes are executed. Server virtualization is no silver bullet

for solving problems in IT operations and management. If server virtualization has been properly

introduced, it can offer faster and better execution of the ITIL processes. The impact is most significant on

the Financial Management process, while also Service Level Management, Incident Management, Change

Management, IT Service Continuity Management and Availability Management are affected considerably.

The impact is less prominent for Application Management, Software Asset Management, Release

Management, Configuration Management and Security Management.

1 INTRODUCTION

Organizations have considered server virtualization

initially from a tactical viewpoint: an effective

technology for consolidation, offering increased

utilization levels, reduced server sprawl, and lower

capital and energy expenses. Over time, server

virtualization is being considered more from a

strategic viewpoint: a catalyst for IT modernization

that changes how IT is acquired, deployed,

consumed, managed, and paid for.

Gartner states that server virtualization is the

highest-impact issue changing IT infrastructure and

operations through 2012 (Dawson & Bittman, 2008).

Gartner also states that server virtualization offers a

natural path to evolve from internal IT

modernization towards cloud computing (Bittman,

2009). Server virtualization enables IT to become

more service-based, allowing scalable and elastic

delivery of resources at much greater speed, driving

economies of scale with shared resources, and

measuring and charging back based on dynamic

usage. Hence, server virtualization makes an IT

organization behave much more like an internal

cloud-computing provider. This paves the way for

outsourcing IT services to external cloud-computing

providers. Once server virtualization has been

introduced, organizations can more accurately

compare internal IT services with external IT

services, and they have gone through fundamental

cultural, political and funding changes that will ease

outsourcing to external cloud-computing providers.

Server virtualization has a clear impact on IT

service management. ITIL (IT Infrastructure

Library) is a set of best practices for IT service

management (Rudd, 2004). Although the impact of

server virtualization on ITIL is widely recognized,

studies on their correlation are still missing. The

research presented in this paper aims to fill this gap.

We first performed a literature study to identify

advantages, disadvantages and risks of server

virtualization, and the correlation between server

virtualization and the ITIL processes. We next

validated and extended these results in an empirical

test by interviewing four organizations that practice

both server virtualization and ITIL. The results of

this research are reported in this paper.

This paper is organized as follows: Section 2

summarizes related work on server virtualization,

ITIL, and their correlation. Section 3 lists the

643

Kamer J. and Vranken H..

THE IMPACT OF SERVER VIRTUALIZATION ON ITIL PROCESSES.

DOI: 10.5220/0003391806430649

In Proceedings of the 1st International Conference on Cloud Computing and Services Science (CLOSER-2011), pages 643-649

ISBN: 978-989-8425-52-2

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

advantages, disadvantages and risks of server

virtualization. In section 4 we correlate these aspects

of server virtualization with the ITIL processes, and

we report the results of an empirical test to validate

this correlation. Section 5 discusses these results and

section 6 concludes the paper.

2 RELATED WORK

ITIL is a set of best practices for IT service

management, focusing on efficiency and cost

effectiveness, and has become a de-facto standard

(Brenner, 2006). ITIL v1 was introduced in 1980.

ITIL v2 appeared in 2000, focusing on service

support and service delivery. ITIL v3 appeared in

2007, arranging processes around the service

lifecycle. Despite the appearance of ITIL v3, many

organizations are still implementing ITIL v2 and

consider this as sufficient or see only limited added

value in ITIL v3 (Pollard & Cater-Steel, 2009). We

therefore only consider ITIL v2 in this paper.

There is plenty of scientific literature on server

virtualization and ITIL, however we discovered only

few sources that discuss the correlation between

them. Furthermore, none of these sources conducted

a thorough scientific study on the correlation.

Mandorla and Hallgårde (2006) state that almost all

IT service management functions are impacted by

the move to server virtualization. Baldwin, Shiu and

Beres (2008) analyse the consequences of server

virtualization on security and audit assurance, and

discuss the impact on some of the ITIL management

processes. Montero (2007) shows the impact of

virtualization on ITIL processes, however without

mentioning how these results were derived.

3 SERVER VIRTUALIZATION

We conducted a literature study to identify

advantages, disadvantages and risks of server

virtualization. Advantages have a positive impact on

the IT organization. Disadvantages have a clear

negative impact, while risks have a possible negative

impact. We identified 34 aspects and grouped them

into five categories: capacity, management,

availability, costs and security (see table 1).

3.1 Capacity

Server virtualization offers better utilization of

existing resources. Multiple virtual servers can be

Table 1: Server virtualization aspects from literature

(Advantage, Disadvantage, Risk).

Capacity

better resource utilization A

difficulty managing resources and peak loads D

additional performance overhead D

Management

increased speed and flexibility of server deployment A

faster deployment of environments for training, test and

development

reduced complexity due to fewer physical servers A

easier backup A

central software update & patch management A

larger consequences of human errors R

unwanted vendor lock-in D

immature management tools and incompatibilities R

not all applications suitable for virtualization D

no support software vendors on virtual environment R

legacy application support A

security concerns when keeping legacy applications R

virtual machine sprawl R

Availability

higher availability levels without additional cost A

faster recovery from crashes and disasters A

reduced application conflicts due to isolation A

hardware defects impacting large number of VMs R

denial of service attack affecting all VMs on host R

Costs

reduced hardware purchases A

increased costs for new, high-end hardware D

reduced management costs A

reduced energy and hosting costs A

increased license costs D

increased costs for personnel training and new hires D

Security

VMM as additional attack vector R

increased security because of VM isolation A

stealing sensitive information R

breaching confidentiality or integrity of VM R

difficult patch management suspended VMs D

secure logging A

rogue virtual appliances R

deployed on a single physical server (Daniels, 2009),

and isolated environments for software testing,

training and development activities can be created

(Kamoun, 2009). However, the response time of

mismanaged virtual machines can become

unpredictable under heavy load (Computer

Associates, 2008). Virtualization also introduces an

additional layer of overhead that must be factored

CLOSER 2011 - International Conference on Cloud Computing and Services Science

644

into the total system load (Loveland et al., 2008).

3.2 Management

Server virtualization can significantly simplify the

task of deploying servers. This often reduces the

provisioning time for a new server from days to

hours. Although tools for managing virtualized

environments are important, they are still immature

(Computer Associates, 2008; Dawson & Bittman,

2008). The number of physical servers is reduced,

thereby reducing the complexity of server

management. Since it is easy to deploy and run

virtual machines, this may result in server sprawl

(Pfister, 2008). Not all applications are suitable to be

virtualized and certain software vendors will not

support their product when virtualized (Tanaka,

Tarui & Naono, 2009; Woltjes & Berg, 2008).

3.3 Availability

Virtualization offers faster disaster recovery

(Montero, 2007). This allows decreased costs and

higher availability levels (Loveland et al., 2008).

However, failure of a physical host affects all virtual

machines running on it (Pfister, 2008).

3.4 Costs

Server virtualization offers reduced costs for

hardware, energy, hosting and management, while

costs for licenses and training are increased.

3.5 Security

Virtualization is implemented by adding a privileged

software layer, which introduces new vulnerabilities.

Attacks against this layer may give access to

multiple virtual machines (Ray & Schultz, 2009). On

the other hand, attacks against a single virtual

machine do not compromise other virtual machines

due to their isolation (Menascé, 2005).

A virtual machine is a set of files, and hence can

be easily copied, inspected or modified, breaching

confidentiality and integrity (Cleeff, Pieters &

Wierenga, 2009). Also patching suspended virtual

machines introduces new challenges (Pfister, 2008).

4 EMPIRICAL RESULTS

Based on literature study, we created a correlation

matrix containing aspects of server virtualization

and ITIL processes. We validated the correlation

matrix in an empirical test consisting of interviews.

Figure 1 shows the conceptual research model.

Figure 1: Conceptual research model.

4.1 Interviews

We interviewed IT managers at the following four

Dutch organizations early July 2010:

 IT department of a university (300 virtual

servers).

 Hosting service provider (500 virtual servers).

 IT organisation of government body on

security and public order (2,300 virtual

servers).

 Large IT consulting firm (8,000 virtual servers

managed internally, thousands at customers).

Before the interviews, the interviewees received a

short briefing on the scope of the research,

definitions of ITIL and server virtualization, and a

description of the 34 server virtualization aspects.

This allowed the interviewees to prepare themselves

for the interview and collect information if required.

The interviews were semi-structured, based upon a

questionnaire that was used in all interviews. The

first questions ask for key figures of the organization

(number of employees, sector and size of IT

organization), how knowledge is acquired, which

ITIL processes are present and which virtualization

types are implemented. Next, the interviewees were

asked to indicate for each aspect of server

virtualization, whether it is applicable in their

organization, how the ITIL processes are impacted,

and whether recommendations can be given. Finally,

the interviewees could identify additional aspects or

recommendations. During the interview, the

interviewees filled an empty correlation matrix. The

interviews were recorded with permission.

4.2 Aspects of Server Virtualization

Table 2 shows the responses of the interviewees

whether they consider each aspect of virtualization

as not applicable, advantage, risk or disadvantage.

Surprisingly, 11 of the 34 aspects are considered as

not applicable by at least 3 interviewees. These

aspects are mainly related to capacity, availability

and security. While these aspects are being

considered in literature as risks or disadvantages, the

THE IMPACT OF SERVER VIRTUALIZATION ON ITIL PROCESSES

645

interviewees indicate that they have been eliminated

by fast improvements of virtualization tools.

Table 2: Server virtualization aspects from interviews (Not

applicable, Advantage, Risk, Disadvantage).

Capacity

better resource utilization 4

difficulty man. resources and peak loads 3 1

additional performance overhead 3 1

Management

incr. speed and flex. of server deployment 4

faster deployment of environments for

training, test and development

1 3

red. complexity by less physical servers 2 1 1

easier backup 1 3

central software update & patch manag. 2 2

larger consequences of human errors 2 1 1

unwanted vendor lock-in 2 2

immature manag. tools and incompat. 1 1 2

not all applications suitable for virt.

3 1

no support software vendors on virt. env.

legacy application support 4

security concerns legacy applications 1 2 1

virtual machine sprawl

2 1 1

Availability

higher av. levels without additional cost 4

faster recovery from crashes and disasters 4

reduced appl. conflicts due to isolation 1 3

hardware defects impacting VMs 4

DoS attack affecting all VMs on host 4

Costs

reduced hardware purchases 1 3

increased costs for new hardware 3 1

reduced management costs 4

reduced energy and hosting costs 1 3

increased license costs 1 2 1

increas. costs for training and new hires 2 2

Security

VMM as additional attack vector 2 2

increas. security because of VM isolation 4

stealing sensitive information 3 1

breach. confidentiality or integrity of VM 3 1

difficult patch manag. suspended VMs 2 2

secure logging 4

rogue virtual appliances 3 1

Table 2 clearly indicates that the interviewees

largely agree on whether aspects are advantages or

disadvantages. They only deviate on the amount of

risk that is still involved with certain aspects. For

instance, one interviewee considers consequences of

human errors as a clear disadvantage, while another

interviewee considers this to be a risk. The

interviewees disagree only on the costs for licensing

and personnel. Although investments for training

and education are generally considered as beneficial,

the dangers are that budgets for doing so are

insufficient and server virtualization is introduced by

unqualified personnel. All interviewees consider

licence costs for virtualization tools as a

disadvantage. However, moving to server

virtualization implies that an organization

reconsiders many IT aspects, including software

licenses. It may be concluded that some software is

no longer required, or can be replaced by other

software. Also, virtualization introduces new licence

structures for operating systems and applications.

The overall net result may be cost saving.

The interviewees also indicated two aspects of

server virtualization that we had not encountered

before. Server virtualization often implies the usage

of SAN (Storage Area Network). Although

maintenance of servers becomes easier with

virtualization, the impact on the IT environment

when doing maintenance on a SAN is very high. The

preferred solution is to implement redundancy,

which however comes at high costs. Another aspect

is managing customer expectations. Thanks to server

virtualization, new servers can be deployed quickly

with high availability. This allows reconsidering

SLAs, but the time available for updating, release

planning, setting up security and solving incidents

becomes more limited. The IT department should

align strategies with the sales department in order to

manage customer expectations.

Table 3 shows the correlation between the

aspects of server virtualization and the ITIL

processes, as indicated by the interviewees. The

numbers in the cells of the matrix indicate the

number of interviewees that identified an impact of a

server virtualization aspect on an ITIL process.

Table 3 clearly indicates that server virtualization

aspects in the categories management and

availability have most impact, while the aspects

related to capacity, costs and security have less

impact. The matrix also indicates that server

virtualization has an impact on all ITIL processes,

especially on the Financial Management process,

while also Service Level Management, Incident

Management, Change Management, IT Service

Continuity Management and Availability

Management are affected considerably. The

following subsections discuss some key findings for

these processes.

The empirical results indicate that server

CLOSER 2011 - International Conference on Cloud Computing and Services Science

646

Table 3: Correlation between aspects of server virtualization and ITIL processes (each matrix cell shows the number of

interviewees that identified an impact of a server virtualization aspect on an ITIL process; totals are sums per row or

column).

Category Aspect

Configuration

Incident

Problem

Change

Release

Service Level

Financial

Capacity

Continuity

Availability

Application

Software Asset

Security

Total

better resource utilization 1 - - 2 2 2 4

difficulty managing resources and peak loads 1 1 1 1

Capacity

additional performance overhead 1 1

increased speed and flexibility of server deployment 2 1 1 4 4 4 2 2 2

faster deploy environm. for training, test and development 2 2 3 2 2 2 1 1 1 1 1

reduced complexity due to fewer physical servers 1 2 2 2 1 1 1 1 1

easier backup 2 2 1 2 1 1 3 1

central software update & patch management 1 1 1 1 2 1 1 1 1 1

larger consequences of human errors 2 1 2 2 1 2 2

unwanted vendor lock-in 1 1 2 1 2 2 2

immature management tools and incompatibilities 1 2 2 1 1 2 2 1 1 1 1 1

not all applications suitable for virtualization 1 1 1 1 1

no support software vendors on virtual environment 2 2 1 2 1 3 1 1 1

legacy application support 1 1 1 1 4 3 1 3 1

security concerns when keeping legacy applications 1 1 2 2 2 1 3

Management

virtual machine sprawl 3 1 2 3 1 1 2

higher availability levels without additional cost 1 1 1 4 4 1 3 4

faster recovery from crashes and disasters 4 1 2 3 1 4 4

reduced application conflicts due to isolation 3 3 2 2 2 1 2 2

hardware defects impacting large number of VMs

Availability

denial of service attack affecting all VMs on host

reduced hardware purchases 1 1 1 2 3 1

increased costs for new, high-end hardware 1 1 1 1 1 1 1 1

reduced management costs 3 2 3 3 1 1

reduced energy and hosting costs 3

increased license costs 1 1 1 4 2

Costs

increased costs for personnel training and new hires 3 1 1

VMM as additional attack vector 1 1 1 1 1

increased security because of VM isolation

stealing sensitive information 1

breaching confidentiality or integrity of VM 1

difficult patch management suspended VMs 1 1 2 1 1 1 1 2

secure logging

Security

rogue virtual appliances 1 1 1 1 1 1 1

Total

16 36 24 35 8 38 50 19 33 32 3 7 17

virtualization hardly influences Application

Management, Software Asset Management and

Release Management. Also the impact on

Configuration Management and Security

Management is limited. Most security concerns

raised in literature, were deemed outdated by the

interviewees and no security breaches due to

virtualization have been reported yet. Configuration

management requires a CMDB of the infrastructure,

storing the mapping of virtual machines to physical

resources and the location of suspended virtual

machines. This is indicated in literature (Baldwin et

al., 2008) and confirmed by the interviewees.

4.3 Financial Management

Financial Management is impacted mostly by server

virtualization aspects in the categories management

and costs. The traditional cost models in IT

organizations are no longer applicable, because

deploying a new virtual server does not necessarily

imply purchasing and charging new hardware. In

order to prevent an explosion in demand of virtual

servers, activity-based costing should be introduced,

which means that usage of services and servers are

billed based on actual resource consumption.

THE IMPACT OF SERVER VIRTUALIZATION ON ITIL PROCESSES

647

Management tools for automatic resource reporting

have not yet matured, so activity-based costing must

still be done manually for now. Server virtualization

also offers higher availability levels, which provides

the opportunity for improved SLAs. It is

inconclusive if virtualization decreases or increases

the total costs.

4.4 Service Level Management

Service levels are much easier to meet when using

virtualization. This is due to an overall increase in

flexibility, and a reduction in downtime and costs.

4.5 Incident Management

The consensus in literature is that Incident

Management becomes more difficult, due to the

added complexity of virtualization which makes

incident solving and root cause analysis more

difficult. The interviewees however clearly indicate

that the advantages of virtualization dominate: the

amount of incidents as well as the time required to

solve them, decreases.

4.6 Change Management

Some authors state in literature that changes are

easier to execute but harder to keep track of (Cleeff

et al., 2009). In practice, certain changes can be

easier because there is no need to acquire new

hardware. Since the risks involved are lower,

changes can be executed more often during daytime.

4.7 IT Service Continuity Management

Virtualization allows much faster recovery from a

crash or disaster, against low costs. There are

however risks involved with virtualization, that

might threaten continuity. Human errors can have

larger consequences in a virtual environment. Also,

some software vendors may not give support if their

products are run in a virtualized environment.

Calamity plans need to be reevaluated for additional

factors that have been introduced by virtualization.

4.8 Availability Management

Virtualization enables high-availability solutions

against very low costs. The number and duration of

downtime incidents will decrease and hardware

failures can be managed better.

5 DISCUSSION

Our research resulted in a comprehensive overview

of ITIL and virtualization implementation aspects

that can for instance be used as a checklist for

organizations that are incorporating ITIL and server

virtualization.

Our results however do not generally apply to all

organizations. We interviewed only a small number

of organizations in a small number of sectors. We

also did not use specific selection criteria for the

interviewed organizations. Therefore, the validity of

our results is limited, and the results of semi-

structured interviews are difficult to reproduce.

Despite these restraints, the interview results contain

little contradictions.

We observed that only 68% of the server

virtualization aspects identified in literature could be

confirmed in our empirical test. It becomes apparent

that server virtualization is rapidly evolving, which

quickly outdates literature. Furthermore, literature

on the correlation between virtualization and ITIL is

largely missing.

Our research could be extended by using a larger

population and applying a quantitative measurable

research method. Future research is also

recommended on security and capacity aspects, and

on measuring costs related to virtualization.

6 CONCLUSIONS

We showed that server virtualization influences all

ITIL processes for IT service support and IT service

delivery. We identified advantages, disadvantages

and risks of server virtualization for capacity,

management, availability, costs and security.

Especially aspects related to management and

availability have a large impact on the ITIL

processes. Our empirical results indicate that some

aspects related to capacity, availability and security,

which are being considered in literature as risks or

disadvantages, have been eliminated by fast

improvements of virtualization tools.

Our main conclusion is that server virtualization

does not change the ITIL processes themselves, but

it does change the way the processes are executed.

The impact is most significant on Financial

Management. Also Service Level Management,

Incident Management, Change Management, IT

Service Continuity Management and Availability

Management are affected considerably.

Server virtualization is no silver bullet for

solving problems in IT service management. If

CLOSER 2011 - International Conference on Cloud Computing and Services Science

648

server virtualization has been properly introduced, it

can offer faster and better execution of the ITIL

processes, and paves the way to cloud computing.

ACKNOWLEDGEMENTS

We kindly thank the interviewees for their

cooperation: Frank Dirkx, Richard van Drimmelen,

Joris Haverkort, Hilde Loozen-Janssen, and Gerhard

van der Zedde.

REFERENCES

Baldwin, A., Shiu, S., & Beres, Y. (2008). Auditing in

shared virtualized environments. Hewlett-Packard

Labs Technical Reports, 4, 2-19.

Bittman, T. J. (2009). Server Virtualization: One Path

That Leads to Cloud Computing. Gartner RAS Core

Research Note G00171730.

Brenner, M. (2006). Classifying ITIL Processes: A

Taxonomy under Tool Support Aspects. Proc. Int.

Workshop on Business-Driven IT Management, 19–28.

Cleeff, A. van, Pieters, W., & Wieringa, R. J. (2009).

Security Implications of Virtualization: A Literature

Study. Centre for Telematics and Information

Technology, University of Twente. Retrieved from

http://doc.utwente.nl/67484

Computer Associates (2008). Virtualization Best

Practices. Retrieved from

http://supportconnectw.ca.com/public/impcd/r11/virtu

alization/doc/virtualization_best%20practices.pdf

Daniels, J. (2009). Server virtualization architecture and

implementation. ACM Crossroads, 16(1), 8-12.

Dawson, P., & Bittman, T. J. (2008). Virtualization

Changes Virtually Everything. Gartner G00156488.

Kamoun, F. (2009). Virtualizing the Datacenter Without

Compromising Server Performance. ACM Ubiquity,

2009(9), 1-11.

Loveland, S., Dow, E. M., LeFevre, F., Beyer, D., &

Chan, P. F. (2008). Leveraging virtualization to

optimize high-availability system configurations. IBM

Systems Journal, 47(4), 591-602.

Mandorla, L., & Hallgårde, F. (2006, November 9).

Leveraging ITIL to Manage Your Virtual

Environment. VMworld Conference 2006, Los

Angeles, USA.

Menascé, D. A. (2005). Virtualization: concepts,

applications, and performance. Conf. Proc. Computer

Measurement Group.

Montero, M. J. (2007). Virtualisatie en IT-auditing.

Postgraduate thesis. Vrije Universiteit, Amsterdam,

The Netherlands.

Pfister, M. (2008). Risk Mitigation in Virtualized Systems.

Master thesis, University of Applied Sciences, Luzern,

Switzerland.

Pollard, C., & Cater-Steel, A. (2009). Justifications,

Strategies, and Critical Success Factors in Successful

ITIL Implementations in U.S. and Australian

Companies: An Exploratory Study. Information

Systems Management, 26(2), 164-174.

Ray, E., & Schultz, E. (2009). Virtualization security.

Proc. 5

Annual Workshop on Cyber Security and

Information Intelligence Research, article 42 .

Rudd, C. (2004). An Introductory Overview of ITIL

(v1.0a). Reading, UK: The IT Service Management

Forum.

Tanaka, T., Tarui, T., & Naono, K. (2009). Investigating

suitability for server virtualization using business

application benchmarks. Proc. Int.Workshop on

Virtualization technologies in distributed computing,

43–50.

Woltjes, & Berg, v. (2008). Criteria voor virtualisatie,

welke zijn relevant? Postgraduate thesis, Vrije

Universiteit, Amsterdam, The Netherlands.

THE IMPACT OF SERVER VIRTUALIZATION ON ITIL PROCESSES

649