期刊名称:International Journal of Computer Science & Information Technology (IJCSIT)
印刷版ISSN:0975-4660
电子版ISSN:0975-3826
出版年度:2012
卷号:4
期号:5
页码:197
出版社:Academy & Industry Research Collaboration Center (AIRCC)
摘要:This paper proposes a novel approach to schedule real-time applications represented by Fork-Join (FJ)task graphs on a cluster platform. The novelty of our approach lies in its ability to efficiently utilize thepower of the cluster’s computational resources to improve the system throughput. Our approach integratestwo heuristic scheduling algorithms: the first algorithm (partition algorithm) works on the cluster level tosearch for the best allocation scheme for the application’s tasks on the cluster’s processors. This search isguided by an objective function that aims to optimize the utilization of cluster’s resources. The secondalgorithm (local scheduler) works on the individual processor level to efficiently utilize the processingpower of each processor. A set of simulation experiments have been conducted to evaluate theperformance of our scheduling approach on both homogeneous and heterogeneous clusters. The resultsshow that our approach improves the acceptance rate of the parallel applications on the cluster comparedto traditional approach.