A Lightweight Model for Estimating Energy Cost of Live Migration of Virtual Machines

Abstract

Live migration, the process of moving a virtual machine (VM) interruption-free between physical hosts is a core concept in modern data centers. Power management strategies use live migration to consolidate services in a cluster environment and switch off underutilized machines to save power. However, most migration models do not consider the energy cost of migration. In a previous study we showed that live migration entails an energy overhead and the size of this overhead varies with the RAM size of the virtual machine and the available network bandwidth. This paper extends our previous work and proposes a lightweight mathematical model to estimate the energy cost of live migration of an idle virtual machine quantitatively. A series of experiments were conducted on KVM to profile the migration time and the power consumption during live migration. Based on these data we derived an energy cost model that predicts the energy overhead of live migration of virtual machines with an accuracy of higher than 90%.