Enabling business-preference-based scheduling of cloud computing resources

Abstract

Although cloud computing technology gets increasingly sophisticated, a resource allocation method still has to be proposed that allows providers to take into consideration the preferences of their customers. The existing engineering-based and economics-based resource allocation methods do not take into account jointly the different objectives that engineers and marketing employees of a cloud provider company follow. This article addresses this issue by presenting the system architecture and, in particular, the business-preferencebased scheduling algorithm that integrates the engineering aspects of resource allocation with the economics aspects of resource allocation. To show the workings of the new business-preference-based scheduling algorithm, which integrates a yield management method and a priority-based scheduling method, a simulation has been performed. The results obtained are compared with results from the First-Come-First-Serve scheduling algorithm. The comparison shows that the proposed scheduling algorithm achieves higher revenue than the engineering-based scheduling algorithm.