An Efficient Virtual Machine Scheduling Algorithm To Minimize Makespan And Maximize Profit Using Hyper Heuristic Approach

Ahmad Alauddin Ariffin, UPM Faculty of Computer Science and Information Technology, Malaysia
2019 International Journal of Advanced Trends in Computer Science and Engineering  
Virtual machines are assigned to hosts, depending on its current resource usage and not considering their overall utilization. It is one of the main problems in cloud computing that can reduces the system performance. The scheduling is used to schedule tasks for better utilization of resources by allocating certain tasks to particular resources at a particular time. The purpose of scheduling is to select the most excellent and suitable resource available to execute the tasks or to assign
more » ... r machines to execute tasks with minimal completion time is but still feasible. An efficient task scheduling algorithm is needed for improve the system performance. In this paper, the focus is on improving the virtual machines scheduling performance for makespan and cost. The proposed process of scheduling includes three main processes. The first process is the Clustering Formation based on the characteristics such as Processor, Memory and Bandwidth. The second process is known as the Hyper Analytical Task Scheduling Algorithm, and based on the scheduled tasks, the Policy-based Profit Maximization Algorithm was proposed in the final process. The performance comparison of the proposed work is analyzed through some empirical results. The result shows that the proposed work significantly reduces the makespan of task scheduling and gives high profit compared with the other scheduling algorithms.
doi:10.30534/ijatcse/2019/3281.42019 fatcat:lfgizkjb6vdplhl44ofaydohgu