Demystifying Cloud Computing
Demystifying Cloud Computing
Demystifying Cloud Computing
Paulo Neto
1 Introduction
2 Overview
In this section we present an overview of Cloud Computing including some def-
initions and a comparison with related concepts.
2.1 Definitions
The underlying concept of cloud computing is not a new one. John McCarthy
[20] in 1961, was the first to publicly suggest (in a speech given to celebrate
MITs centennial) that computer time-sharing technology might lead to a future
in which computing power and even specific applications could be sold through
the utility business model (like water or electricity). This idea of a computer or
information utility was very popular in the late 1960s, but given up by the mid-
1970s as it became clear that the hardware, software and telecommunications
technologies of the time were not ready yet for that challenge. However, since
2000, the idea has resurfaced in new forms [20]. The first academic definition
was provided by [8] in 1997 who called it a computing paradigm where the
boundaries of computing will be determined by economic rationale rather than
technical limits. When Eric Schmidt [19,18] explained his cloud computing view
on the Search Engine Strategies Conference in 2006 and a couple of weeks later
Amazon included the word cloud in the Elastic Cloud Computing (EC2), the
term became a mainstream.
In October 2007 Google and IBM [9,14] announced a major research initiative
to help students and researchers to address this new Internet-scale computing
paradigm. There is still a little consensus how to define Cloud Computing [12].
The Berkeley researchers [16] define Cloud Computing as both the applications
delivered as services over the Internet and the hardware and systems software
in the datacenters that provide those services.
Cloud
Services
Pay as
Elastic Managed
You go
This section presents the cloud computing architecture, the typical deployment
models as well as the underlying technologies.
Application
SaaS
Platform
PaaS
Infrastructure
IaaS
Virtualization
– Internet
– Virtualization
– Software-as-a-Service
– LAMP and WAMP stacks
– Web Hosting
– Database
– Inexpensive CPUs and storage
– SOA (service-oriented architectures)
– Sophisticated client algorithms, including HTML, CSS, AJAX, REST
– Client broadband
– SOA (service-oriented architectures)
– Large infrastructure implementations from Google, Yahoo, Amazon, and
others
Supercomputers
Grids Clouds
Clusters
Web 2.0
Non-Service Service
Applications Oriented
Applications
Although the many advantages of cloud computing already presented, some se-
curity issues need to be carefully evaluated. Processing sensitive data outside the
companies require additional cautions since the data bypass the physical, logi-
cal and personnel controls. The customers need to follow data retention policies
enforced by the regulatory laws, thus the service provider need to be prepared
and certified for those requirements. As long as the data is spread across multi-
ple locations a special concern need to be taken regarding to the law on those
specific locations (state/countries). In order to keep the data confidentiality, a
widely tested encryption scheme should used because an encryption accident can
make the data totally unusable. To protect against theft and or denial-of-service
attacks by users, virtualization is the primary security mechanism in the today’s
clouds. It’s a well known technology that protects against most attempts by
users to attack one another or the underlying cloud infrastructure. Nevertheless
neither all resources are virtualized nor the virtualization engines are bug free.
Another important concern is the protection against the cloud provider. The
virtualization technologies may allow the one who manages the lower layers to
circumvent many security barriers. Although there is already strict legislation in
US and EU to regulate and audit cloud computing providers [17], there is still
some reluctance by several companies to adopt this new technology at least in
a public cloud.
One of the basic policies for any organization is the TCO (Total Cost of Own-
ership) reduction. For instance, provisioning a data center for the peak load
it must sustain during the end of the month (batch jobs), leads to a resource
underutilization during remaining time of the month. An organization in this
situation may benefit from cloud computing paying by the hour for computing
leading to cost savings even if the hourly rate to rent a machine from a cloud
provider is higher than the rate to own one (including space, power, cooling and
maintenance costs).
One of the basic policies for any organization is the reduction the TCO (Total
Cost of Ownership). For instance, provisioning a data center for the peak load
it must sustain during the end of the month leads to underutilization during the
other days. In this situation an organization may benefit from cloud computing
paying by the hour for computing leading to cost savings even if the hourly rate
to rent a machine from a cloud provider is higher than the rate to own one
(including space, power, cooling and maintenance costs).
Another example is a business startup that requires some computing re-
sources at the beginning but will suffer an unexpected increase of the resources
demand when it becomes popular. This type of organization doesn’t need to
invest upfront in computer resources that potentially will be used only months
or years later.
Although the clouding computing costs are more expensive compared with
owning a server for the same period, the cost of over-provisioning (having the
server in almost idle status during many hours per day) or under-provisioning
(increasing the system response time) is very high, driving the cloud computing
solution very attractive from the business point of view.
In Figure 4 with a correctly antecipated peak load, there is a waste of re-
sources due the lack of elasticity. The Figure 5 represents a typical underpro-
visioned environment where the demand overpass the capacity during the peak
load and still presenting long periods of low resource usage. Figure 6 represents
an on demand provisioning where the resources are dynamically adjusted based
on the demand (elasticity) providing a great cost effective solution.
C apacity
Resour ces
Resour ces
C apacity
D emand D emand
0 1 2 3 0 1 2 3
T ime Time
C apacity
D emand
0 1 2 3
T ime
In Figures 4,5 and 6, the grey area represents the waste of resources making
the on-demand provision the most efficient.
Although the economical benefits explained before there are also some nega-
tive aspects using Cloud Computing. For instance transferring a large database
over the network involves a significant cost (high bandwidth) and time. Some
companies choose to ship the data in physical support (tape or disk) through a
courier in order to reduce the network costs and the time that takes the data
import activity.
7 Conclusion
This paper reviewed the Cloud Computing technology from the architecture and
deployment models point of view and presented some economic advantages and
disadvantages and potential security issues. It has been produced as baseline to
start a PhD research within this subject and aims to contribute to the Cloud
computing evolution. Since cloud computing has embedded a blend of technolo-
gies, with so many opened issues like security, it may drive thousands of research
projects that will change the computing paradigms during the current decade.
References
1. Amazon Elastic Compute Cloud (Amazon EC2). http://aws.amazon.com/ec2/,
[Accessed December 2010]
2. Amazon Simple Storage Service (Amazon S3). http://aws.amazon.com/s3/, [Ac-
cessed December 2010]
3. Mint - Free Personal Finance Software. http://www.mint.com/, [Accessed January
2011]
4. The Rackspace Cloud. http://www.rackspacecloud.com/, [Accessed December
2010]
5. Buyya, R., Yeo, C., Venugopal, S., Broberg, J., Brandic, I.: Cloud computing and
emerging IT platforms: Vision, hype, and reality for delivering computing as the
5th utility. Future Generation Computer Systems 25(6), 599–616 (2009)
6. Chaganti, P.: Cloud services for your virtual infrastructure, Part 1: Infrastructure-
as-a-Service (IaaS) and Eucalyptus. IBM developerWorks (December 2009), [Ac-
cessed December 2010]
7. Chaganti, P.: Cloud services for your virtual infrastructure, Part 2: Platform as
a Service (PaaS) and AppScale. IBM developerWorks (January 2010), [Accessed
December 2010]
8. Chellappa, R.: Cloud computing: emerging paradigm for computing. INFORMS
1997 (1997)
9. Chiu, W.: Google and IBM Initiative for Computer Students. http://www.ibm.
com/ibm/ideasfromibm/us/google/images/podcast_google.pdf (October 2007),
[Accessed December 2010]
10. Foster, I., Zhao, Y., Raicu, I., Lu, S.: Cloud computing and grid computing 360-
degree compared. In: Grid Computing Environments Workshop, 2008. GCE’08.
pp. 1–10. Ieee (2009)
11. Furht, B., Escalante, A.: Handbook of Cloud Computing. Springer (2010)
12. Geelan, J.: Twenty-One Experts Define Cloud Computing. http:
//cloudcomputing.sys-con.com/node/612375 (January 2009), [Accessed Decem-
ber 2010]
13. Jha, S., Merzky, A., Fox, G.: Using clouds to provide grids with higher levels of
abstraction and explicit support for usage modes. Concurrency and Computation:
Practice and Experience 21(8), 1087–1108 (2009)
14. Lohr, S.: Google and I.B.M. Join in Cloud Computing Research. The New York
Times (October 2007), [Accessed December 2010]
15. Mell, P., Grance, T.: The NIST Definition of Cloud Computing. National Institute
of Standards and Technology (2009)
16. Michael, A., Armando, F., Rean, G., Anthony, D., Randy, K., Andy, K., Gunho,
L., David, P., Ariel, R., Ion, S., et al.: Above the clouds: A berkeley view of
cloud computing. EECS Department, University of California, Berkeley, Tech. Rep.
UCB/EECS-2009-28 (2009)
17. Sotto, L., Treacy, B., McLellan, M.: Privacy and Data Security Risks in Cloud
Computing. Electronic Commerce & Law Report (2010)
18. Sullivan, D: Conversation with Eric Schmidt hosted by Danny Sullivan.
(http://www.google.com/press/podium/ses2006.html) (August 2006), [Accessed
December 2010]
19. Wikipedia, the free encyclopedia: Eric Schmidt. http://en.wikipedia.org/wiki/
Eric_Schmidt (2010), [Accessed December 2010]
20. Wikipedia, the free encyclopedia: John McCarthy (computer scientist). http:
//en.wikipedia.org/wiki/John_McCarthy_(computer_scientist) (2010), [Ac-
cessed December 2010]
21. Wikipedia, the free encyclopedia: Cloud computing. http://en.wikipedia.org/
wiki/Cloud_computing (2011), [Accessed January 2011]