Research Of Job Schedule System In Cluster And Grid Environment

Because of its outstanding price/performance ratio and scalability, availability and usability, Cluster system is becoming the hotpot and trend of the High Performance Computer Architecture and Parallel Computing research. However, due to rapid development in computer and network communication technology, application requirements evolve towards the trend of high performance, multi-function, heterogeneity and so on. As a cost-effective and flexible solution, grid computing is proposed to fill the need of making good use of hardware and software resources to perform efficient computing operations. In cluster computing and Grid computing environment, job schedule is related to the performance of system, and it is a very important issue.Based on the former works done by others, this dissertation firstly investigates the essential issues of job schedule in cluster and Grid environment, compares some representatives of Cluster Job Management system software and grid scheduling systems. We investigate several popular job-scheduling algorithms, design and implemented a Backfilling based Multi-job scan scheduling algorithm——MS-Backfilling. Then, we propose a cluster job management software which adopt MS-Backfilling scheduling algorithm, and present its design and implementation. The experiment demonstrates that the MS-Backfilling scheduling algorithm can not only solves the problem of 'starvation', but also improves the rate of system utilization, and have better performance compared with normal Backfilling scheduling algorithm. Moreover, the problem of Computational Grid Scheduling is discussed and the fault tolerant grid SPN modeling is proposed. The model is used to have an efficient analysis on the performance of Computational Grid. Finally, we conclude the thesis and the work done by author, and give some aspects of job schedule that we should do next.