Energy-Efficient Algorithms for Dynamic Virtual Machine Consolidation in Cloud Data Centers

Abstract

Cloud computing has become a significant research area in large-scale computing, because it can share globally distributed resources. Cloud computing has evolved with the development of large-scale data centers, including thousands of servers around the world. However, cloud data centers consume vast amounts of electrical energy, contributing to high-operational costs, and carbon dioxide emissions. Dynamic consolidation of virtual machines (VMs) using live migration and putting idle nodes in sleep mode allows cloud providers to optimize resource utilization and reduce energy consumption. However, aggressive VM consolidation may degrade the performance. Therefore, an energy-performance tradeoff between providing high-quality service to customers and reducing power consumption is desired. In this paper, several novel algorithms are proposed for the dynamic consolidation of VMs in cloud data centers. The aim is to improve the utilization of computing resources and reduce energy consumption under SLA constraints regarding CPU, RAM, and bandwidth. The efficiency of the proposed algorithms is validated by conducting extensive simulations. The results of the evaluation clearly show that the proposed algorithms significantly reduce energy consumption while providing a high level of commitment to the SLA. Based on the proposed algorithms, energy consumption can be reduced by up to 28%, and SLA can be improved up to 87% when compared with the benchmark algorithms.