Genetic Algorithm based Scheduling to reduce energy consumption in cloud

Abstract

Cloud computing is the emerging computer technology in the present era. It is an internet based technology and provides shared resources such as databases, storage, servers, softwares to the users as per the demand. Consumers then pay according to the usage. There are various aspects which influence the system performance and scheduling is one of them. So there is a need of an coherent scheduling algorithm that can enhance the overall performance. Scheduling algorithms mainly emphasize on completion time, cost and makespan. Furthermore there are many heuristic based algorithms for scheduling but it is observed that genetic algorithm converges faster and gives optimal results. In this research an efficient scheduling approach is introduced which focuses on reduction of energy consumption. The overall energy which is consumed is determined by resource utilization. The proposed genetic algorithm based approach assigns the task to virtual machine according to the utilization so that overall energy consumption is minimized. Comparison is made between FCFS, SJF and the proposed approach. Results of the evaluation work prove that our approach has least energy consumption among the three algorithms.