Task Scheduling Algorithm based on Multiple-objective Optimal Resource Clustering

ABSTRACT

Resources of grid has the characteristics of large scale, heterogeneous and diverse, in order to improve the efficiency of resource searching, take the way of resources clustering to shorten task and resource matching time. With minimum execution time as the main target, and taking into account the resource load balancing, fusing hypergraph theory, using genetic algorithm for resources multiple-objective optimal clustering pretreated, a task scheduling algorithm based on multiple-objective optimal resource clustering(MORC) is proposed. The simulation results show that this method reduces task and resource matching time in large scale grid computing than Min-min algorithm, and set the resource load threshold, can efficiently balancing resource load and improve the performance of task scheduling at the same time.

Keywords: