Virtual Machine Allocation in Heterogeneous Cloud for Load Balancing Based on Virtual Machine Classification

Abstract

Many organizations are turning to cloud users because of the potential benefits of the cloud. The increasing popularity of cloud services has brought several difficulties as well. Balancing the workload among the available resources at cloud datacenter is one of them and becomes a crucial task. The cloud service provider needs an effective mechanism for achieving workload balance to meet the demands of large numbers of users. To overcome this, many different approaches are suggested in the literature. But still, there is scope to improve the performance of the heterogeneous cloud. The method of distribution of workload among resources needs to consider the processing capability of each resource. Here, in this work, we propose a method “VAHC (VM Allocation in Heterogeneous Cloud for Load Balancing Based on VM Classification)” for allocation of VM based on its classification. The median is used for effective classification of VMs into two groups based on their capacities. This work focuses on minimizing the response time and time required for processing the request in the heterogeneous cloud. The performance of this work is analyzed and compared with “Equally Spread Current Execution (ESCE)”, “Throttled”, and “Round Robin (RR)” Algorithms. The results of the proposed method showed a considerable reduction of 16% in response time whereas 29% in time required processing the request at the datacenter.