An intelligent ensemble learning approach for energy efficient and interference aware dynamic virtual machine consolidation

Inefficient resource usage is one of the greatest causes of high energy consumption in cloud data centers. Virtual Machine (VM) consolidation is an effective method for improving energy related costs and environmental sustainability for modern data centers. While dynamic VM consolidation algorithms can improve energy efficiency, virtualisation technologies cannot guarantee performance isolation between co-located VMs resulting in interference issues. We address the problem by introducing an energy and interference aware VM consolidation algorithm which uses predictive modelling to classify workloads using their resource usage features to make more informed consolidation decisions. The use of ensemble methods plays a pivotal role for improving predictive performance for many different problems. Using recent workload data from Microsoft Azure we present a comparative analysis of several ensemble methods using state-of-the-art prediction models and propose an ensemble based VM consolidation algorithm. Our empirical results demonstrate how our approach improves energy efficiency by 34% while also reducing service violations by 77%.