Energy-efficient communication-aware VM placement in cloud datacenter using hybrid ACO–GWO

Abstract

Virtual machine placement (VMP) is the process of mapping virtual machines to physical machines, which is very important for resource utilization in cloud data centres. As such, VM placement is an NP-class problem, and therefore, researchers have frequently applied meta-heuristics for this. In this study, we applied a hybrid meta-heuristic that combines ant colony optimisation (ACO) and grey wolf optimisation (GWO) to minimise resource wastage, energy consumption, and bandwidth usage. The performance study of the proposed work is conducted on variable number of virtual machines with different resource correlation coefficients. According to the observations, there is 2.85%, 7.61%, 15.78% and 19.41% improvement in power consumption, 26.44%, 57.83%, 77.90% and 83.89% improvement in resource wastage and 2.94%, 8.20%, 9.99% and 10.72% improvement in bandwidth utilisation as compared to multi-objective GA, ACO, FFD and random based algorithm respectively. To study the convergence of the proposed method, it is compared with few recent hybrid meta-heuristic algorithms, namely ACO–PSO, GA–PSO, GA–ACO and GA–GWO which exhibits that the proposed hybrid method converges faster.