Performance Evaluation of Infrastructure as Service Clouds with SLA Constraints

Abstract

Abstract. In this paper, we present an experimental study of job scheduling algorithms in infrastructure as a service type in clouds. We analyze different system service levels which are distinguished by the amount of computing power a customer is guaranteed to receive within a time frame and a price for a processing time unit. We analyze different scenarios for this model. These scenarios combine a single service level with single and parallel machines. We apply our algorithms in the context of executing real workload traces available to HPC community. In order to provide performance comparison, we make a joint analysis of several metrics. A case study is given.