Using COBIT 5 for Risk to Develop Cloud

Computing SLA Evaluation Templates

Onyeka Illoh1,2(&), Shaun Aghili1, and Sergey Butakov1

1
Concordia University of Edmonton,
7128 Ada Boulevard, Edmonton, AB T5B 4E4, Canada
o.illoh@yahoo.com,
{shaun.aghili,sergey.butakov}@concordia.ab.ca
2
Information Systems Assurance Management,
Concordia University of Edmonton, Edmonton, Canada

Abstract. The use of cloud services as a business solution keeps growing, but
there are significant associated risks that must be addressed. Despite the
advantages and disadvantages of cloud computing, service integration and
alignment with existing enterprise architecture remains an ongoing priority.
Typically, quality of services provided is outlined in a service level agreement
(SLA). A deficient template for evaluating, negotiating and selecting cloud
SLAs could result in legal, regulatory, and monetary penalties, in addition to
loss of public confidence and reputation. This research emphasizes (or advo-
cates) the implementation of the proposed SLA evaluation template aimed at
cloud services, based on the COBIT 5 for Risk framework. A gap analysis of
existing SLAs was done to identify loopholes, followed by a resultant template
where identified gaps were addressed.

Keywords: Cloud computing
Cloud users
Cloud providers
Service level

agreements
Software as a service
Platform as a service
Infrastructure as a
service
Everything as a service
COBIT 5 for risk

1 Introduction

1.1 Background
Cloud computing remains a hot topic among vendors, enterprises and end users.
Different authors and industry experts advocate a variety of approaches to realize
benefits at optimal costs, and reduce associated risks from cloud computing [1, 2].
Some of the key benefits include: pay-as-you-go model, scalable solution that supports
rapid business growth, cost transparency to the end user or business, outsourcing of
competencies that are not core to the business, as well as mirrored solutions to mini-
mize the risk of downtime [1, 2].
For users, the cloud computing industry promises tremendous prospects of market
growth, but a wide range of potential risks and safety issues remain prominent [16].
Cloud challenges ranges from data privacy issues, responsibilities for security breach,
loss of physical control, availability concerns, cloud data backup and recovery,
implications for e-discovery, compromised system security, inaccurate billing, greater
dependency on third parties, to the inability of enterprises to satisfy audit/assurance
© Springer International Publishing Switzerland 2015
F. Toumani et al. (Eds.): ICSOC 2014, LNCS 8954, pp. 236–249, 2015.
DOI: 10.1007/978-3-319-22885-3_21
 Using COBIT 5 for Risk to Develop Cloud Computing 237

charter and requirements of regulators or external auditors [1, 2, 18]. Well known
incidents with cloud services include: Amazon’s EC2 cloud service partial outage, the
security breaches of Sony’s PlayStation Network and Qriocity music service [19].
These events emphasized that customers’ inability to control their data remains a key
issue of the cloud computing model [19].
The Institute of Internal Auditors indicated that today’s auditors are faced with
increasingly new-and-improved technologies (including cloud computing) that are
transforming the business environment but introduces new risks that must be managed
[11]. Hence, through this research, an SLA evaluation template aimed at cloud com-
puting services, based on the COBIT 5 for Risk framework was developed.
(a) Cloud Computing Defined
According to the National Institute of Standards and Technology (NIST) and Cloud
Security Alliance (CSA), “Cloud computing is a model for enabling convenient, on-
demand network access to a shared pool of configurable computing resources (e.g.,
networks, servers, storage, applications, and services) that can be rapidly provisioned
and released with minimal management effort or service provider interaction”. Cloud
is composed of five essential characteristics, three service models, and four deployment
models [20, 21].

1.2 Problems with Cloud SLAs

Gartner’s 2010 EXP Worldwide Survey of nearly 1,600 Chief Information Officers,
indicates that spending on IT cloud services is expected to grow almost threefold over
the next five years [14]. Although cloud computing has evolved as a business solution,
there are significant associated risks that must be addressed. As enterprises implement
this technology, the integration and alignment of the services delivered by the Cloud
Service Provider (CSP) remains an ongoing priority.

1.3 Research Questions and Scope

This research attempts to answer these cloud related questions and concerns:
• How can potential cloud users effectively evaluate and select the best suitable CSP
for their business needs while minimizing potential risks (what standard or refer-
ence SLA parameters can be used to measure CSP’s performance)?
• How can cloud users rely on cloud providers to secure and protect their data and
information assets (what assurance do cloud users have about cloud services and
who will provide this assurance)?

2 Discussion and Analysis

2.1 Literature Review

The related works reviewed are categorized into: Cloud Governance, Cloud Computing
Market Maturity, Service level agreements (SLAs), as well as Security, Compliance
and Data Privacy. The various categories are discussed in subsequent sections.
238 O. Illoh et al.

(a) Cloud Governance

According to Gartner, good governance practices, the ability to choose a suitable cloud
computing environment, as well as security and privacy are the key challenges in cloud
computing.1 Boards of Directors are advised to guide the cloud investments to ensure
optimization of risk, control of associated costs and creation of enterprise value [15].
Jirasek highlighted the importance of establishing and enforcing good governance
practices for cloud computing projects [5]. To realize strategic, economic and opera-
tional benefits from cloud computing, enterprise goals and objectives must be aligned
with adoption drivers [3]. Flexibility, reduced initial investment cost, faster deployment
and virtualization are some of the cloud characteristics that may demand more gov-
ernance considerations so that benefits are realized within the enterprise’s risk appetite
[3]. The enterprise’s ability to buy only what it uses was reinforced as one of the goals
of cloud computing [3]. ISACA advised focusing on using competition among cloud
service providers as a bargaining chip to negotiate the best prices since the core can be
provisioned and modified as needed [3].
According to ISACA, in order to discern whether cloud services will meet board’s
expectations, there should be an initial alignment between the enterprise strategy and
expectations [3]. For effective cloud governance, the establishment of a mutual
understanding of expected benefits, as well as tracking and measuring tools should be
prioritized [3]. ISACA proposed COBIT 5 as a tool for governing and managing cloud
investments, in addition to implementing consistent practices to maximize value and
control risk [3]. Hence, it is evident that governance is a key area in cloud computing
that helps ensure alignment with business strategy and priorities. Thus, the importance
of SLA management in the governance framework cannot be overemphasized.
(b) Cloud Computing Market Maturity
Although cloud computing is still in its early stages of maturity, significant concerns
will continue to be addressed [6, 7]. Cloud computing should be seen more as a
business enabler and less as a technology issue so that the technology can progress in
its maturity levels and enterprises can derive promised benefits [7].
The need for executive management to gain an understanding and appreciation for
cloud by seeing it as a source of innovation was also highlighted [7]. Furthermore,
cloud risks should be addressed at the enterprise level rather than as a technical issue
[7]. Hence, to ensure cloud computing progresses through its maturity levels and
benefits are maximized, an enterprise should address risk areas like security, privacy,
data ownership, etc. These can be negotiated with the provider through SLAs.
(c) Security, Compliance and Data Privacy
According to Awad, cloud computing is best implemented through a phased-in
approach [10]. Security, application type to be transitioned, as well as the CSP’s proven
track record, financial stability, and allowance for negotiating suitable terms are the key
considerable factors in selecting the right cloud partner [10].

1
http://www.gartner.com/technology/topics/cloud-computing.jsp.
 Using COBIT 5 for Risk to Develop Cloud Computing 239

Long-term viability, privileged user access, data segregation, recovery, vulnera-

bility to attack or breaches, and regulatory compliance were also among customers’
security concerns listed [8]. Jurisdictional issues, whereby the cloud user needs to
comply with both laws governing its own country and that of the country where its data
is stored was highlighted [8]. Where there is a conflict of laws, further consideration
should be given to security levels at the physical location where cloud services will be
deployed and managed [8]. Since cloud providers are required to physically separate
backup data from production data, location of the offsite backup data becomes a
compliance issue as it could be stored outside the client’s legal/regulatory jurisdictions.
While vendors may not disclose city, state, or country the backup data is stored, they
should be willing to work with the client to provide, and prove compliance of the
offsite backup data location [17].
A survey of IT executives by IDC eXchange, highlighted that security, availability
and performance2 are key challenges facing cloud services. The work done by Sy-
mantec to help enterprises make the right decisions in evaluating CSPs affirmed that
security, compliance and data privacy remain areas of concern when considering the
use of cloud services [12]. This paper focused on how CSPs could leverage secure
sockets layer (SSL) certificates to deliver desired security levels for enterprises. Gartner
Research identified seven areas of security risk associated with cloud computing that
should be evaluated by enterprises when selecting a cloud hosting provider: privileged
user access, compliance, data location, data segregation, recovery, investigative support
and viability [13].
For secure and confidential data, enterprises are under regulations. Outsourcing
services to CSPs does not relinquish consumer’s responsibility for compliance. As due
diligence, cloud users should ensure CSPs are preventing unauthorized third-party
access or modification to address compliance risks [12]. As an added layer of pro-
tection, SSL deployed in backup and recovery ensures that backup data accessed is
encrypted in transit, and servers accessed for backup data are authenticated as legiti-
mate sources [12]. SSL is the proven technology for cloud security as it helps in
developing trust between cloud user and provider [12].
According to Wei, Murugesan, Kuo, Naik and Krizanc, CSPs must implement
strategies that enhance data integrity and privacy to address users’ security concerns
[24]. New data auditing and encryption techniques to protect cloud users’ data from
cyber-attacks while assuring a high level of data availability were also proposed [24].
Hence, addressing security and compliance issues in SLAs remain paramount.
(d) SLAs
According to Gartner, establishing the right SLA prior to the cloud user-vendor rela-
tionship is essential [14]. In a survey, CSA and ISACA affirmed that SLAs form the
basis for clear definition and enforcement of user expectations, in addition to ade-
quately documenting expectations of what the cloud provider will offer [7]. Based on
related work, different but overlapping components or concerns to be addressed in an
SLA are summarized in Table 1.

2
http://blogs.idc.com/ie/?p=730.
240 O. Illoh et al.

Table 1. Summary of existing work related to SLA components

SLA component References
Business requirements for availability, response time for incidents and [1, 4, 7, 9]
additional computing resources, change and patch management
Provisions for disaster recovery, business continuity and physical access [4, 8, 9, 13,
controls 14]
Penalties for non-compliance to SLAs [1, 14]
SLAs for security (Sec-SLAs) [8]
CSP and cloud user’s responsibilities for data and security, alignment of [4, 8, 9, 14]
security metrics with industry standards and practices
Confidentiality agreement, exit strategies, and portability (moving from one [1, 9, 13]
CSP to another)
Data retention and disposal policies and procedures [4, 13]
Controls to satisfy legal, compliance and jurisdictional requirements [4, 8, 13, 24]
Monitoring and performance measurements [1, 8]

Therefore, inadequately defined SLAs contribute to a risky relationship between the

cloud user and service provider [7]. As more enterprises leverage cloud computing,
some of the service providers are offering competitive prices while others are being
distinguished by quality of service through availability, and enhanced security.

2.2 Research Methodology

An initial comprehensive search for existing cloud SLA evaluation templates was
carried out. Next, a list of the ten most important cloud computing companies was
looked at, but the top five CSPs were critically compared, as well as their respective
SLAs: Amazon, VMware, Microsoft, Salesforce and Google [25]. Public cloud vendors
(Amazon, Microsoft and Google) were asked questions related to the infrastructure and
platform services as suggested by cloud analysts and consultants [26]. Note that at the
time of this research, Amazon had created a unique niche market where their larger
cloud consumers are given the opportunity to customize cloud SLAs, thereby defining
terms that best meet their business needs. The comparison is summarized in Table 2
and this is not conclusive due to rapid change of CSP offerings.
The study and analysis by Cloud Spectator compared five large cloud IaaS pro-
viders (Amazon, Rackspace, HP Cloud, SoftLayer and Windows) to determine their
price-performance value, and Microsoft Windows topped the list in terms of customer
satisfaction (performance) [27].
(e) Findings and Results
Cloud SLA concepts adapted from NIST, SLA terms and parameters from ISACA and
CSA were used to formulate a scorecard prototype of SLA components that should be
negotiated and documented in cloud SLAs [4, 21, 22]. According to Tschinkel, security
in addition to data privacy and availability concerns are to be addressed as these are
some of the most critical areas of risk management for the cloud [17]. A vendor market
survey was carried out to justify the selection of cloud providers and at the time of
 Using COBIT 5 for Risk to Develop Cloud Computing 241

Table 2. Top 5 SLAs analysis - side by side comparison

Amazon VMware Microsoft Salesforce Google
Availability/uptime 99.95 % 99.9 % 99.95 % N/A 99.95 %
guarantee
Custom cloud SLA Yes N/A No N/A No
Compensation for Service Service Service N/A Service credit
downtime credit credit credit
Reporting uptime Public N/A Public N/A Public
dashboard dashboard dashboard
Publicly post audits Limited N/A Limited N/A Limited
Audits of controls by No N/A No N/A No
customers/potential
customers
Customer-led penetration Yes N/A Yes N/A No
testing (simulate
cyber-attack)
Response-time to notify Based on N/A Promptly N/A Per
customers of breach applicable contractual
law terms
Customers choose cloud Yes N/A Yes N/A Yes
storage location

research, Amazon and Microsoft emerged at the top for the following reasons: cus-
tomization of cloud SLAs and customer satisfaction. Hence, these best practice SLA
terms were grouped in five areas, though some are applicable in more than one area:
Confidentiality, Integrity, Availability, Auditability and Customer Satisfaction. Case in
point is the ‘Interoperability and Portability’ component that fits into Integrity and
Customer Satisfaction.
Patel, Ranabahu and Sheth proposed the Web Service Level Agreement (WSLA)
framework as a mechanism for managing SLAs in a cloud computing environment, in
addition to being developed for SLA monitoring and enforcement in a Service Oriented
Architecture (SOA) [23]. The third party support feature of WSLA was used to del-
egate monitoring and enforcement tasks, in addition to presenting a real world use case
to validate their proposal [23]. However, a risk-based approach in alignment with
COBIT 5 for Risk was not adapted or mentioned.
At the time of research, no existing cloud computing SLA evaluation template
aligned with the COBIT 5 for Risk framework was found. So, these key SLA terms
were mapped to COBIT 5 for Risk and the resultant scorecard prototype was used to
test Amazon’s SLA.
The scorecard prototype becomes the basis of risk analysis for the IT Risk Sce-
narios in COBIT 5 for Risk framework for cloud SLAs. An IT Risk scenario is an event
that can lead to loss and has a business impact, when and if it occurs [28]. These IT
Scenarios were adapted to the cloud environment and embedded into the gap analysis.
Table 4 shows the mapping of the SLA components to COBIT 5 for Risk. The
complete analysis table shows the twenty example scenarios that were adapted from
242 O. Illoh et al.

Table 3. Testing the scorecard prototype (SLA components) based on best practices from NIST,
ISACA and CSA with Amazon’s SLA
Scorecard prototype Amazon elastic compute
cloud (Amazon EC2) SLA
SLA components Addressed Not Vague
addressed
Confidentiality SSL, Encryption based on data √
classification (data at rest and in
transit)
Data (Information) Dispersion √
Secure Disposal (data security lifecycle) √
User Management, Access √
Control/Authorization
Human Resources/NDAs √
Identity and Access Management √
Segregation of Duties (SoD) √
Third Party Access √
Security controls √
Integrity Interoperability and Portability - must √
not affect data in any way
Data segregation (per multi-tenancy) √
Availability Uptime √
Contingency Planning (IR, DR, BC) √
Data Retention, Backup and Recovery √
Response time √
Source code escrow √
Auditability Independent Audits; sub-categories: √
Type of audit (type I or II), Frequency
(annual/semi-annual), scope/quality
(is CIA covered),
credibility/reputation of the auditing
firm
Change Management, Configuration √
Management and Patch Management
Audit Logging and Monitoring √
Penalty for noncompliance √
Cross-border issues/Compliance with √
Jurisdictional laws on Data Location
Security breach disclosure √
responsibilities
Third party certification (ISO/IEC √
27001/27017, SAS 70, PCI, etc.):
sub-categories-Type, Frequency and
CIA Components should be part of
the report
(Continued)
 Using COBIT 5 for Risk to Develop Cloud Computing 243

Table 3. (Continued)
Scorecard prototype Amazon elastic compute
cloud (Amazon EC2) SLA
SLA components Addressed Not Vague
addressed
System of internal controls (e.g. Policies √
and Procedures)
Review of SLA metrics and compliance √
Right to audit clause √
Customer Pricing Plans √
Satisfaction Performance (usage, load balancing, √
(UnixBench delivery, quality, etc.)
components) Maintenance and Service Support √
Flexibility to Customers’ Request √
Scale Up/Scale Out - Interoperability √
and Portability

COBIT 5 for Risk and tailored to cloud computing. These Risk Scenario Categories are
high level descriptions of the category, while Risk Type are types to which scenarios
derived from the gap analysis will fit (using three risk types which could be primary fit
(higher degree)-P/secondary fit (lower degree)-S/blank for non-related risk scenario).
The three risk types are [28]:
• IT benefit/value enablement risk (resulting from lost opportunities to leverage
technology for new business initiatives or improve the efficiency or effectiveness of
business processes).
• IT programme and project delivery risk (related to the contribution of IT to new or
improved business solutions, through projects and programmes).
• IT operations and service delivery risk (as a result of operational stability, avail-
ability, protection and recoverability of IT services that can destroy or reduce
enterprise value).
Amazon’s publicly available SLA - Elastic Compute Cloud (Amazon EC2) was
tested against the scorecard prototype and results are shown in Table 3 above. The
rating scale in three categories are: Addressed (where the SLA component is clearly
stated), Not Addressed (if not stated) and Vague (if it’s unclear how the SLA com-
ponent is addressed). According to the test, majority of the SLA components fall into
the ‘Not Addressed’ category and are gaps to be discussed or negotiated with the CSP.
This is just an example of how the scorecard prototype can be applied to any SLA.
This initial audit helps in identifying gaps and risks the enterprise needs to manage.
If an SLA component is not stated in the SLA, it becomes the customer’s responsi-
bility. Where the SLA component is important, the cloud consumer should see if it can
be negotiated with the CSP to reduce risk and cost. The importance of evaluating the
amount of risk being shared cannot be overemphasized. This evaluation should also
identify the risk either the consumer or CSP are responsible for. Any risk that cannot be
 Table 4. Mapping of SLA components to COBIT 5 for risk - cloud SLA evaluation template
244

Risk scenarios based on COBIT 5 for Risk

# Risk scenario category Risk Type (P/S) SLA components from Example scenarios
IT IT IT Table 3
benefit/value Program Operations
enablement and project and service
delivery delivery
1 Portfolio Establishment and Maintenance P P S N/A Ensure selected cloud services are aligned with business strategy and
O. Illoh et al.

priorities. Prior to adoption, cloud services should be assessed for

compatibility with existing architecture
2 Program/projects life cycle management P P S Performance (usage, load Cloud projects are within scope, allocated budgets and delivered on time
(program/projects initiation, economics, balancing, delivery, without deteriorating quality. Stakeholders are actively involved from
delivery, quality and termination) quality, etc.) initiation to the end to avoid failure. Change management is deployed to
Change management keep stakeholders informed and future users trained
Maintenance and
Service
Support
3 IT investment decision making P P N/A There’s alignment between business and IT when making cloud investment
decisions. Business case is drawn up to justify cloud investments
4 IT expertise and skills P P P Human Resources/NDAs Due diligence and screening of candidates involved either at the cloud
User Management, consumer’s end or vender’s to ensure appropriate skills and competences in
Access the recruitment process. Security education, training and awareness (SETA)
Control/Authorization to ensure staff are up to date with the latest cloud developments.
Identity and Access Segregation of duties and job rotation to ensure no single employee knows
Management, SoD the entire system. Suitable staff with appropriate skills and competences are
Third Party Access attracted to support business objectives, service and value delivery. Staff
and 3rd party access or authorization is granted based on Least Privilege
and Need-to-Know principles
5 Staff operations (human error and malicious P S P Configuration, Patch, Configuration management is leveraged to elude errors. Avoid authorization
intent) Identity and Access creep so that access rights from prior roles are not abused. Coordination
Management, Uptime, between HR and IT Administration to ensure timely removal of access
Contingency Planning rights to deter abuse
(IR, DR, BC), Backup For security, two separate individuals should approve before actions are taken
and Recovery Response (the 4-eye principle) especially in the areas of backup, information entry,
time, System of internal system maintenance and upgrades. Data centre is secured, monitored for
controls, Security irregularities and only authorized staff are granted access. Ensure
controls Logging appropriate security controls are in place to deter theft. Ensure monitoring
and Monitoring for performance, availability and other irregularities, in addition to prompt
response to alerts
(Continued)
 Table 4. (Continued)
Risk scenarios based on COBIT 5 for Risk
# Risk scenario category Risk Type (P/S) SLA components from Example scenarios
IT IT IT Table 3
benefit/value Program Operations
enablement and project and service
delivery delivery
6 Information (data breach: damage, leakage and P S P SSL, Encryption based on Contingency planning to ensure that if database is corrupted or hardware
access) data classification (data components are damaged, data would be available. Backup procedures
at rest and in transit) based on data classification levels are in place, in addition to testing
Data (Information) backups and protecting backup media. Through continuous network
Dispersion monitoring and firewalls, sensitive information on cloud provider or
Secure Disposal (data consumer’s site is protected. SETA to ensure staff do not accidentally
security lifecycle) disclose sensitive information through social media or email. To protect
Data Retention, data, portable media are secured and encrypted. Intentional modification of
Data segregation (per information is prevented through the 4-eye principle. Regular update of the
multi-tenancy) data retention policy to avoid inefficient archiving, retaining or disposal of
NDAs information. Non-disclosure agreements and intellectual property clauses
Contingency Planning (IR, are factored into contracts to avoid leakage information or trade secrets and
DR, BC), loss of competitive advantage
Backup and Recovery
Security breach disclosure
responsibilities
7 Architecture (architectural vision and design) P S P Scale Up/Scale Out Cloud consumer’s architecture should be flexible to support adoption of newly
acquired cloud services in a timely manner
8 Infrastructure (hardware, operating system and P S P Scale Up/Scale Out Newly acquired cloud services should not make consumer’s systems unstable
controlling technology) Performance (usage, load leading to operational incidents. Underlying infrastructure should allow for
(selection/implementation, operations and balancing, delivery, scale up/scale out in case user volumes increase or handle system load
decommissioning) quality, etc.) when new cloud services are deployed. Cloud services should be tested
prior to deployment into the production environment to ensure system
availability and proper functionality. Hardware and utilities should be
protected from failures, in addition to putting standby measures in place to
support continual execution of critical business transactions
9 Software P S Source code escrow For SaaS models, contingency planning should include source code and data
Change and escrow to assure business continuity regardless of what happens to the
configuration SaaS provider. SaaS customers could enter into an agreement with the 3rd
Management party hosting provider to continue hosting the application in case the SaaS
Using COBIT 5 for Risk to Develop Cloud Computing

Backup and Recovery vendor goes out of business.

(Continued)
245
 Table 4. (Continued)
246

Risk scenarios based on COBIT 5 for Risk

# Risk scenario category Risk Type (P/S) SLA components from Example scenarios
IT IT IT Table 3
benefit/value Program Operations
enablement and project and service
delivery delivery
Change control and change management should be in place to reduce incident
O. Illoh et al.

resolution time and problem management. Roll-back procedures are in place

in case of operational issues, in addition to establishment of backup and
restore points in accordance with business needs
10 Business ownership of IT P S S Review of SLA metrics Cloud initiatives should not be a sole responsibility of the technical team,
and compliance enterprises should assume accountability to ensure alignment with business
strategy. Business requirements should be adequately defined and reviewed
to ensure effective SLAs. Cloud investments are within the procurement
process and weighed based on cost vs. benefits
11 Supplier selection/performance, contractual S P Review of SLA metrics Prior to a strategic partnership, enterprises should exercise due diligence in
compliance, termination of service and and compliance selecting the CSP; check the financial viability, delivery capability, as well
transfer Performance (usage, load as sustainability of the CSP. Cloud services and support should be
balancing, delivery, reviewed to ensure they’re in accordance with the SLA. Defined key
quality, etc.) performance indicators (KPIs) should be linked to rewards and penalties to
Maintenance and Service ensure adequate service delivery and support. If the partnership ceases to
Support, exist, there should be measures that allow for interoperability and
Flexibility to Customers’ portability (exit strategies). To avoid service integration issues with
Request, existing services, the enterprise should consult/involve IT before
Penalty for noncompliance, purchasing cloud services
Interoperability and
Portability
12 Regulatory compliance P S P Independent Audits, Independent audits and 3rd party certification should be carried out to assure
Credibility/reputation of compliance with regulatory standards. The consumer can request for a right
the auditing firm, to audit clause in contractual agreements and ensure the CSP is willing to
Third party certification, work with the consumer to comply with regulations that prohibit
Review of SLA metrics and cross-border dataflow
compliance
Right to audit clause,
Cross-border
issues/Compliance with
Jurisdictional laws on Data
Location
(Continued)
 Table 4. (Continued)
Risk scenarios based on COBIT 5 for Risk
# Risk scenario category Risk Type (P/S) SLA components from Example scenarios
IT IT IT Table 3
benefit/value Program Operations
enablement and project and service
delivery delivery
13 Geopolitical P Cross-border Ensure that compliance to national, support of local initiatives and
issues/Compliance with government interference does not affect the partnership between cloud
Jurisdictional laws on consumers and their service providers, in addition to service capabilities
Data Location
14 Infrastructure theft or Destruction S S P Contingency Planning (IR, Security controls and contingency planning should address theft of servers or
DR, BC) devices with sensitive data, in addition to destruction or sabotage of data
Security controls centres. Access to data centres should be logged, monitored and restricted
Logging and Monitoring only to authorized personnel
15 Malware S P Contingency Planning (IR, Through firewalls, security controls, contingency planning and continuous
DR, BC) monitoring of network, cloud infrastructure should be protected against
Security controls malware, logical bombs, and loss of data through phishing attacks
Logging and Monitoring
16 Logical attacks S P Contingency Planning (IR, Through firewalls, security controls, contingency planning and continuous
DR, BC), monitoring of network, cloud infrastructure should be protected against
Security controls hacking, unauthorized access to systems, industrial espionage and service
Logging and Monitoring interruption due to denial-of-service attacks
Identity and Access
Management
17 Industrial action S S P Contingency Planning (IR, Through contingency planning, alternate solutions where critical business
DR, BC) tasks can be executed should be planned for in case the 3rd party or
primary location becomes unavailable due to strike
18 Environmental S S P Contingency Planning (IR, Ensure that equipment used at data centres is environmentally friendly (e.g.,
DR, BC) power consumption)
19 Acts of nature S S P Contingency Planning (IR, Contingency planning should take into consideration the impact of natural
DR, BC) disasters on cloud services, if and when they occur
20 Innovation P S Security controls New and important technology trends in cloud computing that have been
Pricing plans identified should be timely assessed for business impact and adopted if
required. The security controls and cost of the new trends should be
considered
Using COBIT 5 for Risk to Develop Cloud Computing
247
248 O. Illoh et al.

negotiated with the CSP must be addressed by the consumer through various risk
management practices. The goal is to realize benefits from cloud initiatives while
optimizing resources and managing risks.

3 Conclusion

In this paper, a scorecard prototype was developed to effectively help cloud users
evaluate and select the best suitable CSP for its business needs while minimizing
potential risks. Best practices from NIST, ISACA and CSA were identified as reference
SLA parameters that can be used in SLAs and measurement of provider’s performance.
Incorporating these terms in SLAs (either as standard or negotiated terms), assures
cloud users of their providers’ commitment and responsibility in securing and pro-
tecting their data, as well as information assets.
Though the initial evaluation template has been generalized, this paper is the first in
its direction for future work where each SLA component can be further addressed.
Recommendations for future work also includes taking a company considering moving
to the cloud as a case study, specifically tailoring the template for the company, and
testing the template prior to acquiring cloud services.

Acknowledgement. The first author will like to thank Concordia University of Edmonton’s
research team for their guidance and support in the completion of this work. Their efforts,
knowledge and experience were instrumental in making this paper a success. She acknowledges
the Academic Research Council for the Student Research Grant awarded to her. She is also
thankful to God Almighty, her family and friends; this has been a journey and she is very grateful
for their love, support and encouragement.

References
1. Information Systems Audit and Control [ISACA]: Cloud computing management
audit/assurance program (2010)
2. Gadia, S.: Cloud computing: an auditor’s perspective. ISACA J. 6, 1–2 (2009). http://www.isaca.
org/Journal/Past-Issues/2009/Volume-6/Pages/Cloud-Computing-An-Auditor-s-Perspective1.
aspx
3. ISACA: Cloud governance: questions boards of directors need to ask (2013)
4. ISACA: Security considerations for cloud computing (2012)
5. Jirasek, V.: Cloud governance done right: examples from the trenches. BrightTALK (2013)
6. Sinnett, W.M: In the Cloud and Beyond. Financial Executive (February 2012)
7. CSA and ISACA: Cloud computing market maturity: study results (2012)
8. de Chaves, S. A., Westphall, C.B., Lamin, F.R.: SLA perspective in security management
for cloud computing. In: IEEE ICNS, pp. 212–217 (2010)
9. Subbiah, S., Muthukumaran, S.S., Ramkumar, T.: Enhanced survey and proposal to secure
the data in cloud computing environment. In: IJEST, vol. 5, no. 01 (2013)
10. Awad, R.: Considerations on cloud computing for CPAs. CPA J. 81(9), 11 (2011)
11. Jackson R.A.: Audit in a digital business world. In: The Internal Auditor Magazine, pp. 36–
41 (2013)
 Using COBIT 5 for Risk to Develop Cloud Computing 249

12. Symantec Corporation: Choosing a cloud hosting provider with confidence: Symantec SSL
certificates provide a secure bridge to trusted cloud hosting providers (2012)
13. Heiser, J., Nicolett, M.: Assessing the security risks of cloud computing. Gartner Research,
ID G00157782 (2008)
14. Smith, D.M, Plummer, D.C, Bittman, T.J, Bova, T, Basso, M, Lheureux, B.J, Prentice, B.:
Predicts 2013: cloud computing becomes an integral part of IT. Gartner, ID: G00230929
(2012)
15. Gartner. http://www.gartner.com/technology/topics/cloud-computing.jsp
16. Wu, J., Shen, Q., Wang, T., Zhu, J., Zhang, J.: Recent advances in cloud security.
J. Comput. 6(10), 2156–2163 (2011)
17. Tschinkel, B.: Cloud computing security understanding risk areas and management
techniques (2011)
18. Gordon, M.: The compliant cloud. BrightTALK (2009)
19. Moore, J.: [CNBC]: Reducing security risks in cloud computing. http://www.cnbc.com/id/
43139361/Reducing_Security_Risks_in_Cloud_Computing
20. Badger, L., Grance, T., Patt-Corner, R., Voas. J.: Cloud computing synopsis and
recommendations. In: NIST, vol. 800, p. 146. Special Publication (SP) (2011)
21. CSA: Security guidance for critical areas of focus in cloud computing v3.0 (2011)
22. NIST: NIST US government cloud computing technology roadmap, Release 1.0 (Draft) - In:
NIST, vol. 500, p. 293. Special Publication (SP) (2011)
23. Patel, P., Ranabahu, A., Sheth, A.P.: Service level agreement in cloud computing (2009)
24. Wei, D.S.L., Murugesan, S., Kuo, S., Naik, K., Krizanc, D.: Enhancing data integrity and
privacy in the cloud: an agenda. IEEE Comput. Soc. 46, 87–90 (2013)
25. Bort, J.: The 10 most important companies in cloud computing. Business Insider (2013)
26. Loftus, T.: Public cloud vendors side by side by side. Wall Street J. 1–3 (2013). http://blogs.
wsj.com/cio/2013/02/26/public-cloud-vendors-side-by-side-by-side/
27. Cloud Spectator: Cloud server performance: a comparative analysis of 5 large cloud IaaS
providers (2013)
28. ISACA: COBIT 5 for risk framework, pp. 67–74 (2013)