Multiple QoS Guided Heuristic for Independent Task Scheduling in Grid

Abstract

Scheduling a task with multiple QoS needs such as deadline, reliability, cost, trust, etc., is called QoS based scheduling. Several heuristics have been proposed for QoS based scheduling and it has been proved that it is a NP-hard problem. In this paper, we have proposed a heuristic for multiple QoS based scheduling for independent tasks. The heuristic considers multiple QoS needs of a task and finds the total utility of the task. It groups the tasks on the basis of their utility values. It schedules tasks in the descendent order from higher to lower utility values. The results show that the proposed heuristic is better in makespan and load balancing than other heuristics such as QoS Guided Min-Min and Weighted Mean Time Min-Min Max-Min Selective. © Springer-Verlag Berlin Heidelberg 2010.