Joint virtual machine placement and migration scheme for datacenters

Abstract

Efficiently managing virtual machines (VMs) plays a key role in improving the resource usage and power consumption of datacenters. However, most of the existing techniques have been optimized for each system criterion separately (e.g., either platform layer or virtualization layer). Such approaches are simple to implement, but they usually result in poor system performance or under-utilization of the resources. In this paper, we investigate the problem of optimal VM management (i.e., placement and migration) in datacenters via a multi-objective function. We show that optimizing the proposed multi-objective function reduces not only the energy consumption, but also the cross network traffic among platforms. Due to the high complexity of the combinatorial optimization problem, we propose an efficient heuristic algorithm to find a near optimal solution based on the relaxed convex optimization framework. We provide both theoretical analysis and simulations to show the effectiveness of the proposed approach over the existing approaches.