Clustered Server Consolidation for Energy Efficient Cloud

Abstract

Energy efficiency for cloud clusters requires a better allocation of resources. With the goal of energy-wise optimization in clouds, we aim to minimize the number of active servers on clouds by migrating VMs without exceeding any server's memory capacity. We introduced a hierarchical algorithm and compared its efficiency against the implementations of Best-Fit Greedy (BFG), Worst-Fit Greedy (WFG), and Simulated Annealing (SA). Our algorithm performed better in almost all scenarios for small, medium, and large-size problems, even with time limitations.