Virtual machine power metering and its applications

Abstract

With the rapid growth of power consumption and the cooling cost in data centers, it is more and more important to improve the utilization rate of power by more sophisticate power monitoring and management strategy. VM power consumption measurement and related scheduling strategy contribute to make cloud computing center more efficient on the basis of making full use of existing power and cooling resources. In this paper, we propose methods to meter the power consumption of VM, and design scheduling strategies based on power consumption to migrate VM between hosts in order to achieve power capping. Specifically, we propose an improved VM power metering method, in which the power consumption of cooling fans is isolated. Besides, we solve the problem of the dynamic power metering where the hyper-threading of host processor is enabled. In addition, we propose VM migration and scheduling strategy on the basis of power consumption. Last but not the least, we propose a strategy to monitor data center status, including power consumption and resource utilization of all hosts and VMs. All above are demonstrated an evaluated in a prototype system.