Energy Efficient and High Performance Scheduling Algorithm for Grid Computing

Abstract

Grid computing is an epitome that furnishes seamless access to widely distributed resources for solving the big computing applications. Large amount of power consumption in grid computing infrastructure is the main concern because it conducing high operational costs and carbon emission to the environment. To reduce the power consumption without degrading the performance is the main challenge of grid computing. In this paper we present a scheduling algorithm which focuses on simultaneous optimization of both energy consumption and performance. We take an energy aware layered architecture in which Task analyzer (layer 1) analyses & sort the tasks on the basis of their complexities and dependencies then Provisioning Manager (layer 2) provisions the resources and tasks on the basis of energy and performance. Then Scheduler (layer 3) schedules the tasks on the resources in accordance of EEHP algorithm. We have validated our approach by conducting an experiment using the Gridsim toolkit. The results demonstrate that a large amount of energy can be saved without degrading the performance.