树形异构网格的启发任务调度算法

Heuristic Tasks Scheduling on Heterogeneous Tree-Based Grid Computing Platforms

讨论了在网格资源计算能力和网络通信速度异构的树形网格环境下任务调度问题,导出了线性方程并且根据调度任务大小进行了模型的优化,提出一个基于线性规划的任务分配启发式算法。实验结果表明:在异构树形计算网格环境下实现任务调度时,该算法的性能明显优于其他算法。

This paper firstly discussed the problem of scheduling independent tasks on tree-based heterogeneous grid computing platforms. Then presented an integer linear programming model for this problem. A heuristic model is also presented. At last, A demand-driven and dynamic heuristic algorithms is proposed. The experimental results show that the algorithms for the scheduling problem obtain better performance than other algorithms.