An Energy Efficient VM Migration Algorithm in Data Centers

Abstract

In order to reduce operation and maintenance expense and also increase the resource utilization rate, server consolidation and virtualization solutions have been adopted in modern cloud computing data centers. Further, the scheduling policies of virtual machine (VM) migration have been regarded as an effective method for energy conservation. In this paper, we address the problem of VM consolidation in cloud data centers. A power aware scheduling algorithm THR_MUG, which is combined with a utilization threshold strategy and a VM selection policy, is proposed. THR_MUG tries to select the most appropriate VMs for migration each time when a physical machine (PM) is considered as being overloaded, such that the utilization of this PM is just not more than the utilization threshold, such that both the energy consumption and the number of VM migration can be reduced. The experimental results show that compared with other algorithm, the proposed algorithm can effectively reduce the number of VM migrations as well as the energy consumption.