Abstract
Wide-ranging edge cloud data centers are a vital part of the solution for the problems caused by enormous growth in the IT industry for high computational power by advanced service applications. Majority of IoT applications switched to the Cloud and this stimulated the emergence of Edge technology to better manage the computing applications, data, resource and services. Consequently, with the massive client size and enormous applications trying to benefit from the cloud service, it makes it a challenging task for the edge cloud data centers to work in a power saving mode. In this paper, we propose a virtual machine consolidation method to switch the idle physical servers into hibernation mode, resulting in reduced power usage. We know that edge cloud data centers offer storage as a service, in this study we address the issues pertaining to storage units in the data centers. A unique classification approach is adopted to ensure load is balanced accordingly during allocation and our main contribution is on the VM migration technique. The VM migration is aimed at consolidating the VMs based on the workload to reduced number of physical machines to mitigate the energy consumption and promoting green computing. Therefore, we name the approach as Workload Aware Virtual Machine Consolidation Method (WAVMCM). We validate the proposed method with a competitive analysis of experimental results gathered from comparing it with Artificial Intelligence based probabilistic algorithm like Simulated Annealing, Genetic Algorithm and a case of no migration. Experimental results demonstrate that the proposed WAVMCM reduces 9% active servers saving 15% of power consumption when compared to genetic algorithm based method. (C) 2018 Elsevier Inc. All rights reserved.