Optimized energy and SLA-aware virtual machine placement strategies in Cloud: study

Abstract

Pay-as-you-go is the latest trend for most of the business applications and to suit it precisely a process should be followed which might be the Service Level Agreement (SLA). Cloud computing requires a clear agreed SLA signed by the service consumer and committed by a service provider. The efficiency of a datacenter depends a lot on how virtual machines are provisioned and where they are located. Virtual machine placement is an optimization problem aiming for multiple goals. An efficient VM allocation policy will improve energy efficiency while limiting the degradation of the quality of service (QoS) and alleviate hotspots, but will also reduce the operating costs of the data Center. In this paper, we study strategies of optimal VM allocation policy to minimize power consumption in a data center while preserving QoS. CloudSim simulator is used to create a cloud environment. We evaluate and compare our algorithms corresponding to different approaches in order to find the one that optimizes VM placement. The simulation result shows that some virtual machine placement strategies minimize the energy consumption and SLA violations.