Multi-objective Meta-heuristic Technique for Energy Efficient Virtual Machine Placement in Cloud Data Centers

Abstract

Cloud computing has emerged as an efficient scalable solution for storing and processing a large amount of data.  Cloud data centers provide the resources on demand to the consumers on pay-per-use model. However, the large amount of datacenters is required to support the growing demand of the cloud consumers. This should be handled in an optimized way to avoid the resource wastage so that more consumers can get the benefits of the data centers. Virtualization is the technology of creating virtual version of the computers called Virtual Machines (VM). A Virtual Machine Placement problem is a fundamental challenge in cloud computing where the goal is to determine the optimal allocation of Virtual Machines to Physical Machines (PM) within a data center. An efficient Virtual Machine Placement technique helps to properly place the VMs on the PMs which significantly optimize the number of servers, maintenance cost, CPU utilization and power consumption. We present a novel hybrid approach that combines Ant Colony Optimization (ACO) algorithm and Sine Cosine Algorithm (SCA) for an efficient VM placement. Since SCA is an emerging search algorithm using Sine and Cosine functions in Engineering field, it has been used to explore the solutions obtained by ACO algorithm  ACO algorithm has been applied to exploit the solutions of the search space for efficient VM placement which helps in power management and also to minimize the resource wastage. The result has been verified with the performance against other algorithms to prove that our proposed algorithm outperforms the other algorithms