Data Center Selection Based on Cloud Customers’ Preferences of Quality Attributes for Federated Clouds

Data Center Selection Based on Cloud Customers’ Preferences of Quality Attributes for Federated Clouds

Maryam Pourreza, Hamid Haghshenas, Jafar Habibi

Abstract

Evolution of cloud computing has introduced a new paradigm called Federated Clouds. Using this concept, even small cloud providers can collaborate with other cloud providers and offer a much huger pool of resources with a much higher quality. Finding an optimized solution for the data center selection problem in a federated environment can help all cloud providers to have an increase in their customers’ satisfaction. Therefore, in this paper we investigate the data center selection problem for federated clouds with the aggregated architecture in order to enhance the satisfaction of all cloud providers’ customers in this environment. So, we first propose an optimization problem which considers all cloud customers’ quality requirements along with their own priorities. We also show that this problem is NP-hard and inspect the complexity of different variations of the proposed problem. Furthermore, five algorithms are introduced for solving the proposed problem. Evaluation of these algorithms reveals that the proposed Ant Colony System algorithm provides the highest satisfaction for the customers in a reasonable amount of time. Moreover, the overall customers’ utilities obtained by using this algorithm is at least 1.232 times higher than those obtained in previous approaches in this context.

Keywords

Data Center Selection, Federated Clouds, Aggregated Architecture, Cloud Computing

References