Energy Efficient Virtual Machine Placement With an Improved Ant Colony Optimization Over Data Center Networks

Abstract

Virtualization has achieved widespread adoption in various fields. As one of the most significant technologies in network virtualization, an effective virtual machine placement can facilitate to improve resource utilization in data center networks and cut down the enterprises' operation cost. In this paper, we proposed an energy efficient virtual machine placement scheme, which pursued to reduce communication cost and power consumption over traffic-aware data center networks. To solve such an optimization problem, an improved ant colony optimization with adaptive parameter setting was presented to balance its fast convergence and robust search capability. Compared with existing algorithms, simulation results demonstrated that our scheme achieved improvements in power consumption and communication cost, and had a significant reduction in run time under different traffic patterns and configurations.