 An article authored by George Darzanos, Jordan Koutsopoulos and George D. Stamoulis (AUEB) has been accepted to be published in the proceedings of the 4th IEEE International Conference on Cloud Networking (CLOUDNET 2015). In the arctile, the problem of the formation of an economically sustainable computational resource federation by Cloud Service Providers (CSPs) is considered. The federation aims at providing improved quality of service in terms of average total delay per job for customers served by the CSPs. Each CSP is modeled as an M/M/1 queue. The queue may serve requests coming from that CSP’s own customers as well as requests coming from customers of other CSPs. Our model includes all salient factors to evaluate the economics of such a federation. In particular, we explicitly model the energy consumption cost of each CSP as function of the resource utilization factor. The CSP’s revenue is modeled by a delay-dependent pricing function with which each CSP charges its clients. We propose a model for the formation and evaluation of a service-based cloud federation according to which each CSP may transfer a portion of the requested workload from its clients to other CSPs within the federation. A federation policy is specified by the portions of workload transferred from each CSP towards other CSPs. We formulate the problem of finding the federation policy that maximizes the total profit (revenue minus cost) of CSPs and we introduce a rule for the fair sharing of the profit that is induced by the federation. Finally, we conduct several experiments under different setups. The numerical results show that by employing our model CSPs can maximize the total profit of the federation and achieves a QoS that approaches the optimal one. |
News >