Authors:
Lucas Borges de Moraes
1
;
Pedro Cirne
2
;
Fernando Matos
3
;
Rafael Stubs Parpinelli
1
and
Adriano Fiorese
1
Affiliations:
1
Santa Catarina State University (UDESC), Brazil
;
2
Instituto de Telecomunicações, Portugal
;
3
Federal University of Paraíba (UFPB)
Keyword(s):
Cloud Computing Provider, Data Envelopment Analysis, Performance Indicator, Selection Method, Ranking.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence and Decision Support Systems
;
Enterprise Information Systems
;
Information Systems Analysis and Specification
;
Internet Agents
;
Modeling of Distributed Systems
;
Software Agents and Internet Computing
;
Strategic Decision Support Systems
Abstract:
Cloud computing become a successful service model that allows hosting and distribution of computational resources all around the world, via Internet and on demand. This success leveraged and popularized its adoption into all major IT companies. Based on this success, a large number of new companies were competitively created as providers of cloud computing services. This fact can difficult the customer ability to choose among those several cloud providers the most appropriate one to attend their requirements and computing needs. Therefore, this work aims to propose a model capable of selecting and ranking cloud providers according the analysis on the most efficient ones using a popular Multicriteria Decision Analysis (MCDA) method called Data Envelopment Analysis (DEA), that calculates efficiency using Linear Programming (LP) techniques. To accomplish that, the efficiency modeling is based on the analysis of each Performance Indicator (PI) values desired by the customer and the avail
able ones in the cloud provider database. An example of the method’s usage is given to illustrate the model operation, selection results and final provider ranking for five hypothetical customer requests and for ten providers.
(More)