An Energy-SLA Aware VM Selection Algorithm for Dynamic Virtual Machine Consolidation

Abstract

Energy efficiency is an essential issue in cloud computing. It has a massive financial and environmental effect that needs attention. A good VM-host mapping achieves low energy consumption while minimizing the number of migrations and Service Level Agreement Violation (SLAV). Dynamic Virtual Machine Consolidation (DVMC) is an excellent solution to reduce Energy Consumption (EC). However, an aggressive DVMC may increase the SLAV. So there is a need to balance the trade-off between EC and SLAV. Therefore, This work presents a Knapsack-based VM Selection (KVMS) algorithm. It works on the dynamic 0/1 knapsack approach and selects the VM with the maximum ratio of its CPU utilization and migration time. Simulation results prove that proposed KVMS reduces SLAV and Energy-SLAV (ESV) at most 60% and 64%, respectively.