OPTIMAL TASK SCHEDULING IN THE CLOUD ENVIRONMENT USING A MEAN GREY WOLF OPTIMIZATION ALGORITHM

Cloud computing is one of the emerging areas in computing platforms, supporting heterogeneous, parallel and distributed environments. An important challenging issue in cloud computing is task scheduling, which directly influences system performance and its efficiency. The primary objective of task scheduling involves scheduling tasks related to resources and minimizing the time span of the schedule. In this study, we propose a Modified Mean Grey Wolf Optimization (MGWO) algorithm to enhance system performance, and consequently reduce scheduling issues. The main objective of this method is focused upon minimizing the makespan (execution time) and energy consumption. These two objective functions are elaborated in the algorithm in order to suitably regulate the quality of results based on response, in order to achieve a near optimal solution. The implementation results of the proposed algorithm are evaluated using the CloudSim toolkit for standard workloads (normal and uniform). The advantage of the proposed method is evident from the simulation results, which show a comprehensive reduction in makespan and energy consumption. The outcomes of these results show that the proposed Mean GWO algorithm achieves a 8.85% makespan improvement compared to the PSO algorithm, and 3.09% compared to the standard GWO algorithm for the normal dataset. In addition, the proposed algorithm achieves 9.05% and 9.2% improvement in energy conservation compared to the PSO and standard GWO algorithms for the uniform dataset, respectively.