See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/278117154

Cloud Computing: Issues & Challenges

Conference Paper · December 2013

CITATIONS READS
16 3,509

2 authors:

Mohammad Sajid Zahid Raza

Aligarh Muslim University Jawaharlal Nehru University
12 PUBLICATIONS   66 CITATIONS    52 PUBLICATIONS   226 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Mohammad Sajid on 13 June 2015.

The user has requested enhancement of the downloaded file.

 International Conference on Cloud, Big Data and Trust 2013, Nov 13-15, RGPV

Cloud Computing: Issues & Challenges

Mohammad Sajid Zahid Raza
School of Computer and Systems Sciences School of Computer and Systems Sciences
Jawaharlal Nehru University Jawaharlal Nehru University
New Delhi, India New Delhi, India
[email protected] [email protected]

Abstract—Cloud computing is a multifarious technological important industry”. Douglas Parkhill first explored the
paradigm that is outgrowth of decades of research in parallel characteristics of cloud computing in 1966 in his book “The
computing, virtualization, networking and communication, utility Challenge of the Computer Utility”. Cloud computing can be
computing and Service-Oriented Architecture (SOA). It offers an thought as an extension of Virtual Private Network (VPN) over
on-demand and scalable access to a shared pool of resources
hosted in a data centre at providers’ site. It reduces the overheads
network infrastructure which is used in telecommunication
of up-front investments and financial risks for the end-user. The world. Initially, telecommunication service providers delivered
qualitative services and lower cost of services are the key dedicated point-to-point circuit which was the wastage of the
requirements of this technology. Regardless of the fact that cloud bandwidth; the problem was solved by using VPN services
computing offers great advantages to the end users, there are where traffics could be switched to balance the utilization of
several challenging issues that are mandatory to be addressed. the overall network. Cloud computing was a buzz word for
Owing to the financial nature of use of the cloud services based on many years and it turned into reality in 2007 when IT giants
Service Level Agreements (SLA) makes these issues even more Google and IBM announced a collaboration in this domain
serious that needs to be taken care of. This work presents an followed by “Blue Cloud” announcement by IBM [2, 3, 4].
overview, style and actuality of cloud computing with the
objective of presenting challenging issues concerned with various
According to blog [5], the prediction of IT advisory company
aspects of cloud computing. Gartner says that cloud computing business will surpass $148
billion mark by 2014 while its competitor, Forrester, says it
Index Terms— Cloud Computing, IT, IaaS, PaaS, SaaS. will reach $118 billion. Another Gartner’s Survey says that the
investment on services in public cloud is expected to increase
I. INTRODUCTION 18.6% in 2012 to $110.3B that achieves a 17.7% Compound
The advancements in Information Technology (IT) demand Annual Growth Rate (CAGR) from 2011 through 2016 [1]. In
a new computing paradigm that supports delivery of computing general, the total market is likely to increase to $210B in 2016
services on minimal charges without installing them at local from $76.9B in 2010. Figure 1 gives a glimpse of the
sites. Cloud computing offers the same model having above distribution of workloads in cloud and a traditional data centre
describe properties in which services are delivered over internet that shows that popularity of cloud will be grow with a very
in an on-demand elastic way for which the charges are paid at fast rate. Therefore, cloud computing area looks very
release time of resources. In general, cloud is a multifarious promising for researchers and businesses. On the other hand,
technological paradigm that is an extension of many existing its realization brings many challenging issues that need to be
technologies viz. parallel and distributed computing, Service- carefully addressed.
Oriented-Architecture (SOA), virtualization, networking etc.
The distributed computing, virtualization and internet works as The organization of remaining paper is as follows. Section
indispensable building blocks of the cloud computing. It is a 2 presents an overview of cloud computing, its essential
highly sharable computing paradigm where processing, storage, characteristics, different deployment models and service
network, applications etc. are shared. The objective of the cloud models. Section 3 describes the advantages and disadvantages
computing is to provide secure, qualitative, scalable, quick, of cloud. Section 4 describes various issues and challenges of
more responsive, on demand, cost-efficient and automatically cloud computing that are necessary to address in order to adopt
provisioned services viz. computation services, storage this technology. Finally, section 5 concludes the papers.
services, networking etc. being provided in a transparent way
(location independent). Cloud computing can help to improve
business performance while making a contribution to control
the cost of delivering IT resources to any organization.
The fundamental idea of cloud computing was pronounced
way back in 1960 by Professor John McCarthy, as; “If
computers of the kind I have advocated become the computers
of the future, then computing may someday be organized as a
public utility just as the telephone system is a public utility.
The computer utility could become the basis of a new and Fig.1.Workload distribution in cloud and traditional data centers [1]
 International Conference on Cloud, Big Data and Trust 2013, Nov 13-15, RGPV

II. CLOUD COMPUTING web enabled devices viz. computers, laptops, mobile
To properly understand cloud, it is important to know what phones etc.
it is, some essential characteristics that a system must possess  Resource pooling: The resources that can be assigned to
to qualify as a cloud along with various services that can be users can be processing, software, storage, virtual
offered using it through various deployment models. machines and network bandwidth. The resources are
pooled to serve the users at a single physical location
A. The Definition and/or at different physical location according to the
Cloud computing is in its infant form and numerous optimality conditions (e.g. security, performance,
definitions have been proposed by many scientists. Some of consumer demand). The cloud gives an impression of
the definitions are, resource location independence at lower level (e.g.
Buyya et al. defines, “A Cloud is a type of parallel and server, core) but not at the higher level (e.g. datacenter,
distributed system consisting of a collection of inter-connected city, country).
and virtualized computers that are dynamically provisioned  Rapid elasticity: The beauty of cloud computing is its
and presented as one or more unified computing resource(s) elasticity. The resources appear to users as indefinite and
based on service-level agreements established through are also accessible in any quantity at any time. The
negotiation between the service provider and consumers” [6]. resources can be provisioned without service provider
intervention and can be quickly scale in and scale out
According to P. Gaw, “The way I understand it, ‘cloud according to the user needs in a secure way to deliver
computing’ refers to the bigger picture…basically the broad high quality services.
concept of using the internet to allow people to access
 Measured service: A metering capability is deployed in
technology-enabled services. According to Gartner, those
cloud system in order to charge users. The users can
services must be 'massively scalable' to qualify as true 'cloud
achieve the different quality of services at different
computing' [1].
charges in order to optimized resources at different level
J. Kaplan says, “I view cloud computing as a broad array of of abstraction suitable to the services (e.g. SaaS, PaaS
web-based services aimed at allowing users to obtain a wide and IaaS).
range of functional capabilities on a 'pay-as-you-go' basis that C. Service Models
previously required tremendous hardware/software
The cloud services are delivered in three forms viz.
investments and professional skills to acquire. Cloud
Infrastructure-as-a-Service (IaaS), Software-as-a-Service
computing is the realization of the earlier ideals of utility
(SaaS) and Platform-as-a-Service (PaaS). The services are
computing without the technical complexities or complicated
delivered over the network by using Web browser, Web Based
deployment worries”[7].
mail etc. The service models are as follows:
The National Institute of Standards and Technology (NIST)
 Software-as-a-Service (SaaS): In this multitenant
defines, “A model for enabling convenient, on-demand
service model, the consumers use application running on
network access to a shared pool of configurable computing
a cloud infrastructure. The cloud infrastructure including
resources (e.g., networks, servers, storage, applications, and
(servers, OS, Network or application etc.) is managed
services) that can be rapidly provisioned and released with
and controlled by the service provider with the user not
minimal management effort or service provider interaction.
having any control over the infrastructure [8, 9]. Some
This cloud model promotes availability and is composed of
of the popular examples are SalesForce.com, NetSuite,
five essential characteristics, three service models, and four
IBM, Microsoft and Oracle etc.
deployment models” [8].
 Platform-as-a-Service (PaaS): With this model, the
B. Five Essential Characteristics provider delivers to user a platform including all the
The cloud computing must have some characteristics in systems and environments comprising software
order to meet expected user requirements and to provide development life cycle viz. testing, deploying, required
qualitative services. According to NIST [8], these five tools and applications. The user does not have any
essential characteristics can be classified as: control over network, servers, operating system and
storage but it can manage and control the deployed
 On-demand self-service: A consumer can access application and hosting environments configurations [8,
different services viz. computing capabilities, storage 9]. Some popular PaaS providers are GAE, Microsoft’s
services, software services etc. as needed automatically Azure etc.
without service provider’s intervention.
 Infrastructure-as-a-Service (IaaS): In this service model,
 Broad network access: To avail cloud computing the provider delivers to user the infrastructure over the
services, internet works as a backbone of cloud internet. With this model, the user is able to deploy and
computing. All services are available over the network run various software’s including system or application
and are also accessible through standard protocols using softwares. The user has the ability to provision
computing power, storage, networks. The consumers

36
 International Conference on Cloud, Big Data and Trust 2013, Nov 13-15, RGPV

have control over operating systems, deployed A. Advantages

applications, storage and partial control over network. Cloud computing offers many benefits and flexibility to its
The consumer has no control over underlying users. User can operate from anywhere at any time in a secure
infrastructure [8, 10]. Some important IaaS providers are way. With the increasing number of web-enabled devices used
GoGrid, Flexiscale, Joyent, Rackspace etc. now-a-days (e.g. tablets, smart phones etc.), access to one’s
D. Deployment models: information and data must be quick and easier. Some of these
relevant benefits in respect to the usage of a cloud can be as
Cloud systems can be deployed in four forms viz. private,
follows [9, 10]:
public, community and hybrid cloud as per the access allowed
to the users and are classified as follows:  Reduces up-front investment, Total Cost of
Ownership (TCO), Total Operational Cost (TOC) and
 Private cloud: This deployment model is implemented
minimizes business risks.
solely for an organization and is exclusively used by
their employees at organizational level and is managed  Provides a dynamic infrastructure that provides
and controlled by the organization or third party. The reduced cost and improved services with less
cloud infrastructure in this model is installed on premise development and maintenance cost.
or off premise. In this deployment model, management  Provides on-demand, flexible, scalable, improved and
and maintenance are easier, security is very high and adaptable services on pay-as-you go model.
organization has more control over the infrastructure and
accessibility [8, 10].  Provides consistent availability and performance with
automatically provisioned peak loads.
 Public cloud: This deployment model is implemented
for general users. It is managed and controlled by an  Can recover rapidly and has improved restore
organization selling cloud services. The users can be capabilities for improved business resiliency.
charged for the time duration they use the services.  Provides unlimited processing, storage, networking
Public clouds are more vulnerable to security threats etc. in an elastic way.
than other cloud models because all the application and
 Offers automatic software updates, Improved
data remains publicly available to all users making it
more prone to malicious attacks. The services on public Document Format Compatibility and improved
cloud are provided by proper authentication [8, 10]. compatibility between different operating systems.
 Community cloud: This cloud model is implemented  Offers easy group collaboration i.e. flexibility to its
jointly by many organizations with shared concerns viz. users on global scale to work on the same project.
security requirements, mission, and policy  Offers increased return on investment of existing
considerations. This cloud is managed by one or more assets, freeing capital to deploy strategically.
involved organizations and can be managed by third
 Provides environment friendly computing as it only
party. The infrastructure may exist on premise to one of
uses the server space required by the application
the involved organization or it may exist off premise to
which in turn reduces the carbon footprints.
all organizations [8, 10].
 Hybrid cloud: This deployment model is an B. Disadvantages of Cloud Computing
amalgamation of two or more clouds (private, Every coin has two faces. That’s not to say, of course,
community, public or hybrid). The participating clouds cloud computing is without disadvantages. Some of the
are bound together by some standard protocols. It disadvantages while using a cloud can be summarized as [9]:
enables the involved organization to serve its needs in  Requires high speed network and connectivity
their own private cloud and if some critical needs (cloud
constantly.
bursting for load-balancing) occur they can avail public
cloud services [8, 10].  Privacy and security is not good. The data and
application on a public cloud might not be very
III. ADVANTAGES AND DISADVANTAGES OF CLOUD secure.
COMPUTING
 Disastrous situation are unavoidable and recovery is
The degree of acceptance for any computing paradigm is not possible always. If the cloud loses one’s data, the
measured by its strengths and weaknesses. If advantages owing
user and the service provider both gets into serious
to computing paradigm are fair enough and the overheads are
bearable to some extent, the degree of acceptance is very high problems.
and that computing paradigm will be accepted by users.  Users have external dependency for mission critical
Following are some important advantages and disadvantages applications.
offered by the cloud can be written as:  Requires constantly monitoring and enforcement of
service level agreements (SLAs).

37
 International Conference on Cloud, Big Data and Trust 2013, Nov 13-15, RGPV

IV. ISSUES AND CHALLENGES OF CLOUD COMPUTING A. Security and Privacy

The existing computing paradigms viz. distributed According to the survey of International Data Corporation
computing, SOA, networking etc. are building blocks of cloud (IDC), Security, Performance and Availability are the three
computing. There are numerous issues associated with theses biggest issues in cloud adoption. The critical challenge is how
computing paradigms and some new challenges emerged from it addresses security and privacy issues which occur due to
cloud computing are required to be addressed properly in order movement of data and application on networks, loss of control
to realize the cloud to its full extent. Current cloud adoption is on data, heterogeneous nature of resources and various security
associated with numerous challenges as shown in Figure 2 and policies. Data stored, processing and movement of data outside
3 depicting the specific business risk of adopting cloud the controls of an organization poses an inherent risk and
services and biggest barriers. Therefore, these issues must be making it vulnerable to various attacks. The security threats
addressed in order to provide high quality services to the users can be of two types viz. internal and external. The external risk
while complying with the service provider’s needs. The issues is posed by various persons and organizations e.g. enemies or
can be organized into several different categories varying from hackers that do not have direct access to the cloud. The internal
security, protection, identity management, resource security risk is a well-known issue which can be posed by
management, power and energy management, data isolation, organizational affiliates, contractors, current or former
availability of resources, heterogeneity of resources. Although, employees and other parties that have received access to an
there are several issues that demand attention but the following organization’s servers, networks and data to facilitate
could be treated as of prime concern [11-14]: operations. Cloud computing poses privacy concerns because
the service providers may access the data that is on the cloud
that could accidentally or deliberately be changed or even
removed posing serious business trust and legal consequences
[8, 11-14].
B. Performance
According to IDC’s survey, performance is the second
biggest issue in cloud adoption. The cloud must provide
improved performance when a user moves to cloud computing
infrastructure. Performance is generally measured by
capabilities of applications running on the cloud system. Poor
performance can be caused by lack of proper resources viz.
disk space, limited bandwidth, lower CPU speed, memory,
network connections etc. Many times users prefer to use
services from more than one cloud where some applications
are located on private clouds while some other data or
applications being on public and/or community cloud. The
data intensive applications are more challenging to provide
proper resources. Poor performance can results in end of
service delivery, loss of customers, reduce bottom line
Fig. 2. Specific Business Risk of adopting cloud services [14] revenues etc. [2, 11, 13].
C. Reliability and Availability
Any technology’s strength is measured by its degree of
reliability and availability. Reliability denotes how often
resources are available without disruption (loss of data, code
reset during execution) and how often they fail. One of the
important aspect that creates serious problems for the
reliability of cloud computing is down time. One way to
achieve reliability is redundant resource utilization.
Availability can be understood as the possibility of obtaining
the resources whenever they are needed with the consideration
to the time it takes for these resources to be provisioned.
Regardless of employing architectures having attributes for
high reliability and availability, the services in cloud
computing can experience denial of service attacks,
performance slowdowns, equipment outages and natural
disasters. Data shows that some of the current cloud computing
providers have some frequent outages last year. e.g Amazon
Fig. 3. Biggest Barriers to adoption of cloud services [14]

38
 International Conference on Cloud, Big Data and Trust 2013, Nov 13-15, RGPV

EC2 outage. In order to remove FUDD (fear, uncertainty, resource utilization. Since cloud computing is a combination
doubt, and disinformation), probably the reliability, availability of many existing technologies, existing job scheduling
and security are the important and prime concern to an strategies are eligible to be applied on cloud system. The major
organization. Therefore, the level of reliability and availability issues of job scheduling on cloud systems are partitioning of
of cloud resources must be considered as a serious issue into jobs into parallel tasks, interconnection network between
the organization’s planning to set up the cloud infrastructure in clouds or processors, assigning priority to jobs and selection of
order to provide effective services to consumers [19]. processors or cloud to allocate the job(s), job flexibility, level
of pre-emption supported, workload characteristics, memory
D. Scalability and Elasticity
allocation, task execution monitoring, recourse allocation
Scalability and elasticity are the most amazing and unique requirements, topology, nature of the job, effect of existing
features of the cloud computing. These features provide users load, load balancing, parallelism, job migration policy,
to use cloud resources being provisioned as per their need in redundant Resource selection, synchronization,
unlimited amount as required. Scalability can be defined as the communication overheads, job pre processing requirements
ability of the system to perform well even when the resources etc. The job scheduling is one of critical process that must be
have been scaled up. Elasticity, on the other hand, is the ability decided very carefully and wrong selection of scheduling
to scale resources both up and down as and when required. strategy can lead to devastating effect on performance leading
Elasticity goes one step further, though, and does also allow to wastage of resources while falling to meet Quality of
the dynamic integration and extraction of physical resources to Service (QoS) standards.
the infrastructure. The elastic cloud computing means that
allocation of resources can get bigger or smaller depending on G. Energy Consumption
the requirement. Elasticity enables scalability—which means According to a survey done by Amazon as shown in Figure
the system can easily scale up or down the level of services to 4, the cost consumption of Amazon data centers is shocking as
which the user has subscribed. Scalability can be provided in 53% of the total cost is consumed by the servers for a 3-year
two ways- horizontally and vertically whereby horizontal amortization period while energy and cooling requirements
scalability (Scale Out) refers to addition of more nodes to the consume 42% of the total budget including both direct power
system such as adding a new computer to an existing service consumption (~19%) and the cooling requirements (23%) for
provider system while vertical scalability (scale up) refers to amortization period of 15-years [16]. In 2006, data centers of
addition of resources to a single node in the system, typically United States consumed more than 1.5% of the total energy
involving the addition of memory or processors to a single produced in that year, and this percentage is expected to
computer [19]. increase 18% annually [15].
E. Interoperability and Portability
Interoperability is the ability to use the same tools or
application across various cloud service providers platforms.
The interoperability can be defined at various levels viz.
application, service, management and Data interoperability.
Cloud users must have the flexibility of migrating in and out
and switching to clouds whenever they want without no vendor
lock-in period. One of the adoption barriers in cloud
computing interoperability is the vendor lock-in risk. The main
problems to realize it are the lack of open standards, open APIs
and lack of standard interfaces for VM formats and service
Fig. 4. Monthly server, Power and Infrastructure costs [16]
deployment interfaces. Cloud portability ensures that one
cloud solution will be able to work with other platforms and Cloud data centers house thousands of servers and set up
applications as well as with other clouds [14]. the cooling infrastructure to remove heats generated by these
servers. These servers and cooling infrastructure consume a
F. Resource Management and Scheduling
large amount of energy and produces green house gases
Resources management can be consider at various levels (GHGs). In addition, the cloud data centres which are inherent
viz. hardware, software, virtualization level with performance, part of the cloud infrastructure are also very expensive to
security and other parameters being dependent on the operate and consume energy at a very large scale. For example,
management and provisioning of resources. It includes the the power consumption of Google data centre is equivalent to a
management of memory, disk space, CPU’s, cores, threads, city such as San Francisco. Since ICT aids towards developing
VM images, I/O devices etc. Resource provisioning can be applications and facilities for human prosperity, we require
defined as allocation and management of resources to provide designing such hardware, software, scheduling policies,
desired level of services. Job scheduling is a type of resource networks and other protocols that consume energy in eco-
provisioning where jobs execution order is established in order friendly and optimized manner. The goal is not only to reduce
to finish job execution to optimize some parameters viz. the consumption of energy and hence the cost consumed by
turnaround time, response time, waiting time, throughput and

39
 International Conference on Cloud, Big Data and Trust 2013, Nov 13-15, RGPV

data centers, but also to maintain environmental standards Deployment (16.8%), Server Capacity (15.6%), and Storage
necessary not only to survive but to thrive [16, 20]. Capacity (15.5%). This survey, with a sample size of 244,
reveals that organizations still are not ready to move their data
H. Virtualization
on to cloud due to security/privacy concerns. The survey also
Virtualization is the creation of a virtual version of a shows that peripheral functions such as IT management and
storage device, an operating system, a server or network personal applications are the easiest IT systems. The
resources. The virtualization divides the resource into multiple organizations are conservative in employing IaaS compared to
execution environments and hides the physical characteristics SaaS being ready to outsource their marginal functions to
of computing resources to simplify the way in which other cloud while critical activities being kept domestic. The survey
systems, applications or end users interact with those also reveals that the organization will move their 31.5%
resources. Virtualization is used in two forms viz. Type 1 Storage Capacity to the cloud in next 3 years while the
hypervisors / Bare-Metal Virtualization and Type 2 collaborative applications will reach to 46.3% by that time
hypervisors / OS virtualization. Virtualization is one of the [21].
key technology in order to make it possible to realize the cloud
computing. Virtualization realization typically enables However, various issues and challenges of cloud
consumers to migrate their computation and data to a remote computing have been taken up in this section still there are
location with some varying impact on performance. There are many other compelling issues that needs to be considered.
numerous benefits of virtualization which could not otherwise Some of these like capacity planning, management of
be achieved. Some of the benefits include elasticity and additional and remaining resources, management of
scalability, cost effectiveness, location independence, automation of resources, costing model, Service Level
infrastructure independency, customization, simplified access Agreement (SLA) etc. are also there demanding an early
interface etc. However, virtualization provides many benefits attention. These issues should not be considered as road blocks
to users, while on the other hand it poses many challenges to in the pursuit of cloud computing, it is rather important to give
cloud computing. It has many critical issues to be address viz. serious consideration to these issues and explore the possible
VM sprawl challenges, workload characterization of VMs, ways out before adopting this technology.
security issues in hypervisor based cloud communication, Live V. CONCLUSION
migration security, unnecessary migration to a private cloud
etc. Virtualization makes infrastructure management more Cloud computing can be considered as an integral
complex, and massive automation is required in order to component of almost all businesses in near future and it is
support the key aspects such as automation, on-demand and expected to change the landscape of IT industry. It is based on
elasticity requirements. the model of delivering services on internet with pay-as-you-
go model with advantages like no up-front cost, lower IT staff,
I. Bandwidth Cost lower cost of operation to name a few. Although cloud
High speed communication channels work as a backbone computing has bright prospects both for business and
of cloud computing. With cloud computing, business gets the researchers certain challenging issues including security,
ability to save money on hardware and/or software but still performance, reliability, scalability, interoperability,
requires spending more on the bandwidth. It is almost virtualization etc. needs to be addressed carefully. The
impossible to fully exploit the services of cloud computing improvement in bandwidth technology, corresponding service
without high speed communication channels. Migration to models and security models can really revolutionize this area
cloud almost removes the up-front cost, while it increases the along with the IT industry. The paper has discussed the
cost of data communication on network i.e. the cost involved concept of cloud computing and shades some lights on various
in transfer of data to and from the private and other clouds issues and challenges that needs to be addressed in order to
[17]. This problem is prominent if consumer application is data realize the implementation of the cloud and making it a
intensive and the consumer’s data is distributed amongst a dominant part of our life in order to thrive.
number of clouds (private/public/community). Cloud
References
computing provides lesser cost for CPU intensive jobs than
[1] E. Anderson et al., Forecast overview: Public cloud services,
data intensive jobs with gray’s argument “Put the computation worldwide, 2011-2016, 4Q12 Update, Gartner Inc., February
near the data” still applicable for data intensive jobs still 2013.
finding relevance [18]. In other words, data intensive [2] IBM, Google and IBM announced university initiative to
applications can perform better being employed on private address internet-scale computing challenges, October-2007.
cloud rather than public/hybrid cloud. [3] S. Lohr, Google and I.B.M. join in Cloud computing
research, October-2007.
J. What to Migrate [4] IBM, “IBM introduces ready-to-Use Cloud computing”,
According to a Survey conducted by IDC in 2008, seven IT November-2007.
systems/applications being migrated to the cloud are: IT [5] SherWeb, Where is the cloud headed in 2013, January-2013,
Management Applications (26.2%), Collaborative Available at http://blog.sherweb.com/where-is-the-cloud-
headed-in-2013.
Applications (25.4%), Personal Applications (25%), Business
Applications (23.4%), Applications Development and

40
 International Conference on Cloud, Big Data and Trust 2013, Nov 13-15, RGPV

[6] R. Buyya, C. S. Yeo, S. Venugopal, J. Broberg and I.

Brandic, “Cloud computing and emerging IT platforms:
Vision, hype, and reality for delivering computing as the 5th
utility”, Future generation computer systems, vol. 25, no. 6,
pp. 599–616, June 2009.
[7] J. Geelan, Twenty one experts define cloud computing,
August 2008. Article available at http://virtualization.sys-
con.com/node/612375.
[8] L. Badger, T Grance, R. P. Comer and J. Voas, DRAFT
cloud computing synopsis and recommendations,
Recommendations of National Institute of Standards and
Technology (NIST), May-2012.
[9] D. A. Menasce and P. Ngo, “Understanding cloud
computing: Experimentation and capacity planning,” in Proc.
of computer measurement group conf., pp. 1-11, December
2009.
[10] IBM Global Services, Cloud computing: defined and
demystified explore public, private and hybrid cloud
approaches to help accelerate innovative business solutions,
April-2009.
[11] Y. Ghanam, J. Ferreira and F. Maurer, “Emerging issues &
challenges in Cloud- A hybrid approach”, Journal of
software engineering and applications, vol. 5, no. 11, pp.
923-937, November 2012.
[12] M. A. Vouk, “Cloud computing – Issues, research and
implementations”, Journal of computing and information
technology, Vol. 16, no. 4, pp. 235-246, June- 2008.
[13] T. Dillon, C. Wu and E. Chang, “Cloud computing: Issues
and challenges”, 24th IEEE International Conference on
Advanced Information Networking and Applications. pp. 27-
33, 2010.
[14] European CIO Cloud Survey, Addressing security, risk and
transition, May -2011.
[15] Energy Star, Report to congress on server and data centre
energy efficiency, Public Law 109-431, U.S. Environmental
Protection Agency, August -2007.
[16] J. Hamilton, “Cooperative expendable micro-slice servers
(CEMS): Low cost, low power servers for internet-scale
services”, Proc. 4th Biennial Conf. Innovative Data Systems
Research (CIDR), Jan 2009.
[17] A. Leinwand, The hidden cost of the cloud: Bandwidth
charges, July 2009.
[18] J. Gray, "Distributed computing economics," ACM Queue,
vol. 6, pp. 63-68, May 2008.
[19] G. T. Lepakshi, “Achieving availability, elasticity and
reliability of the data access in cloud computing”, Int. journal
of advanced engineering sciences and technologies, vol 5,
no. 2, pp. 150-155, April 2011.
[20] A. Berl et al., “Energy-efficient cloud computing,” The
Computer Journal, vol. 53, no. 7, pp. 1045 – 1051, 2009.
[21] F. Gens, New IDC IT cloud services survey: Top benefits
and challenges, IDC exchange, February 2010.

41

View publication stats