Abstract:
Grid computing has become one of the most popular topics in both research community and commercial companies. Allocation policy is one of the most important factors, which affect the performance of Grid Computing system. A new Grid Computing System based on an optimized resource allocation algorithm was proposed in this paper. It contains four components: Grid User, Grid Broker, Grid Resource and Grid Information Service. The optimized resource allocation algorithm, consisting of user side resource allocation and resource side job scheduling, was emphasized in this paper. With the combination of both sides' efforts, the jobs will be allocated to the most appropriate resources and completed in the most suitable order to meet the requirements of users. At last, the experiments and results comparison with traditional particle swarm optimization and Max-min scheduling are presented, which proves that the proposed algorithm has better performance than the other two algorithms.