HBAC Algorithm for Enhancement of Makespan and improved Task allocation for VM in cloud datacenter

Abstract

Regardless of the past research work in cloud computing some of the challenges still exist related to workload distribution in cloud data centers. Especially in the infrastructure as a service IaaS cloud model. Efficient task allocation is a crucial process in cloud data center due to the restricted number of resource and virtual machines (VM). IaaS is one of the main models of cloud computing because this model handles the backend where servicer like VM and data centers are managed. Cloud service providers should ensure high service delivery performance in such models avoiding situations such as hosts being overloaded or under loaded as this result causes VM failure and make higher network execution time. Therefore, to overcome these problems, this paper proposed an improved load balancing technique known as the HBAC algorithm which dynamically allocates resources by hybridizing the Artificial Bee Colony (ABC) algorithm with the Bat algorithm. The proposed HBAC algorithm was tested and compared with other state-of-the-art algorithms on 200-20000 even tasks by using CloudSim on standard workload format (SWF) data sets file size (200kb and 400kb). The proposed HBAC showed an improved accuracy rate in task distribution of VM in a cloud datacenter and reduced the makespan (energy level) in the datacenter. Based on the ANOVA comparison test results, a 1.98 percent improvement on accuracy or task distribution of VM occurs and 0.98 percent reduced makespan or energy level of cloud data center. The results are consistent with different services broker policies which are used during simulation process for the proposed algorithm in cloud datacenter. In future research the proposed algorithm used for predication approach for resource managements system in cloud data center.