A modified grouping-based job scheduling in computational grid

Abstract

Grid computing provides a high performance computing platform to solve applications with large number of independent jobs. However, user jobs developed for grid might be small and of varying lengths according to their computational needs and other requirements. In fact, it is a big challenge to design an efficient scheduler, but there exists some grouping based job scheduling strategy that intends to minimize total processing time by reducing overhead time and computation time, and on the other hand maximizing resource utilization than without grouping based scheduling. Further analysis and research on job scheduling can be carried out to enhance the performance of grouping based scheduling algorithm in grid computing. This study intends to achieve better performance by extending the concept of grouping based job scheduling. Therefore, this paper proposes “A modified grouping-based job scheduling in computational grid” with the objective of minimizing overhead time and computation time, thus reducing overall processing time of jobs. The work is verified through simulation and the results obtained shows that the proposed grouping-based scheduling algorithm is better than, others.