基于资源结点性能预测的启发式调度算法研究  

Research on heuristic scheduling algorithm based on resource node performance prediction

作　　者：魏妮妮 宋翌[1] 

机构地区：[1]武汉生物工程学院计算机与信息工程系,湖北武汉430415

出　　处：《实验技术与管理》2013年第10期51-54,共4页Experimental Technology and Management

基　　金：湖北省自然科学基金项目(2011CD297);武汉市教育局科研项目(2009k101)

摘　　要：网络的任务调度策略的好坏将直接影响网格系统的性能。在分析已有启发式任务调度算法的基础上,设计了一个资源节点性能预测模型,在此模型基础上提出了一种基于资源结点性能预测的任务调度算法。仿真实验结果表明,该算法可使任务的总的完成时间减小,当处理任务量相同的情况下,能使得系统资源的利用率得到提高,这表明本算法是有效的,更适用于高吞吐量应用任务调度问题的解决。Grid task scheduling is based on certain optimization objective as the guiding ideology ,scheduling strategy will directly affect the performance of grid system . Based on analysing the existing heuristic scheduling algorithm ,a resource node performance prediction model is designed .Based on this model a task scheduling algorithm with resource node performance prediction is proposed .With a simulation experiment , the results show that when the task is under the same conditions ,this algorithm not only makes the overall task completion time reduce ,but also improves the utilization ratio of system resources ,so it is proven that the algorithm is effective ,and more suitable for applications in high-throughput scheduling .

关 键 词：网格计算 任务调度 性能预测 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]