How to Choose the Right Cloud - Advanced Design Approaches to Emerging Software Systems - page 238

total cost of ownership is reduced and mainte-
nance burden is shifted to providers. On the other
hand on premise supporters argue about security,
systems' redundancy, functionality and data pri-
vacy as obstacles to cloud computing. Aspects
that can point out the way to IT deployment are
potential costs, user demand and desired quality
attributes. User demand is an indicator of the load
of a system and the estimated traffic that greatly
affects infrastructure costs. A thorough five year
cost analysis will enlighten potential long term
cost benefits of both solutions. Desired quality
attributes on the other hand and the level these
attributes are incorporated into on premise and on
cloud solutions can also affect the final decision.
Today, most organizations tend to adopt ex-
clusively one of the two solutions limiting the
possibilities that a combined solution can offer.
An hybrid approach can provide the best of both
worlds by allowing the customer organizations to
maximize the benefits of both a hosted delivery
model and those of the on-premise model. Such
a model may exploit just IaaS combined with on
premise software applications to avoid infrastruc-
ture costs. An alternative is to use SaaS on VPNs
to minimize potential data privacy risks. Or even a
company can use PaaS service to build each own
applications and deploy them using IaaS or private
infrastructure. Services offered by the cloud cover
a wide variety of IT needs. A potential customer
can find the optimal development and deployment
solution keeping in mind all relevant aspects of
his own specific IT problem and how these are
incorporated in the two models.
Closing, as future work we aim to evaluate
the proposed model on real world applications
deployment and compare the three alternatives
(cloud, on-premise, a combination of the two)
based on data coming from both in-house devel-
opment and cloud hosting. In particular for the
hybrid of the two worlds, we plan to elaborate on
the cases where it is more profitable and derive
appropriate “rules-of-thumb”, since we argue

that this model will be the one that will finally
dominate the market.

REFERENCES

Aggarwal, S. (2005). TCO of on-demand ap-
plications is significantly better for SMBs and
mid-market enterprises. Yankees Group report.
Retrieved March 10, 2010 from http://www.
intente.net/pdfs/ Yankee_On_Demand_vs_On
_Premises_TCO_1_.pdf?ID=13165

Albrecht, A. J. (1979). Measuring application
development productivity. Proceedings of the
Joint SHARE, GUIDE, and IBM Application
Development Symposium , (pp. 83-92). Monterey,
California, October 14-17, IBM Corporation.

Amazon Elastic Cloud. (2010). Amazon platform
as a service . Retrieved March 10, 2010, from
http://aws.amazon.com/ec2/

Armbrust, M., Fox, A., Griffith, R., Joseph, A. D.,
Katz, R. H., Konwinski, A., et al. (2008, Febru-
ary). Above the clouds: A Berkeley view of cloud
computing . (Technical Report EECS-2009-28),
University of California at Berkeley.

Boehm, B. (1981). Software engineering econom-
ics . Englewood Cliffs, NJ: Prentice-Hall.

Cloud Computing Congress. (2010). Cloud com-
puting China . Retrieved March 10, 2010, from
http://www.cloudcomputingchina.org/

Dikaiakos, M. D., Katsaros, D., Mehra, P., Pallis,
G., & Vakali, A. (2009). Cloud computing: Dis-
tributed Internet computing for IT and scientific
research. IEEE Internet Computing , 13 (5), 10-13.
doi:10.1109/MIC.2009.103

Gray, J. (2003, March). Distributed computing
economics . (Technical Report MSR-TR-2003-24),
Microsoft Research.