Adaptive Hybrid Optimization Based Virtual Machine Placement in Cloud Computing

Abstract

The rapid evolution of cloud technology has attracted millions and millions of users due to its unlimited services and the cloud service providers have put huge efforts to construct the large scale data centers. The data center is a repository of Physical machines (PMs) and large-scale of Virtual Machines (VMs) are utilized to reduce resource cost of PMs. Typically, cloud computing is accomplished only by means of virtualization technology and the mechanism of allocating VMs to consumers is called VM placement. However, VM placement is performed by considering various parameters, such as allocation time, SLA violation, energy consumption, resource utilization, and so on. The VM placement issue is a major constraint for good VM consolidation and the ultimate intention of this VM placement issue is to reduce quantity of functioning host machines in cloud by alleviating energy consumption and increasing the resource usage. Therefore, an effective mechanism is needed to effectively balance the network by addressing the aforementioned issues. In this research, an effective strategy called Adaptive-Artificial Bee Cat Swarm Optimization (ABCSO) algorithm is developed for VM optimal positioning in cloud environment. The developed Adaptive-ABCSO is derived by the integration of adaptive concept into the classical ABCSO algorithm. The traditional ABCSO algorithm is obtained by combining Artificial Bee Colony (ABC) and Cat Swarm Optimization (CSO). Moreover, the proposed approach has achieved minimum load of 0.165, minimum migration of 0.055 cost and low power consumption of 0.042.