摘要： 针对执行时间限制严格的DAG类型网格工作流任务调度问题，考虑到网格环境中存在多个性能相同的网格资源，但其有效度和价格各不相同将会对工作流任务调度产生影响，该文利用有限状态连续时间的Markov过程的数学模型，提出一种网格工作流调度算法。在DAG中的关键路径上资源系统有效度满足用户要求的一定信任水平，选择执行费用相对较低的资源。仿真实验结果验证了算法的有 效性。

关键词: 工作流调度, 关键路径, 关键区间有效度

Abstract: Aiming at the workflow scheduling of DAG-based workflow on the grid in which time constraints are rigid, considering that the number of grid resource is very likely more than one and that these resources with different reliability and different service price can finish the certain task within the same time limit, the paper presents a stochastic algorithm in terms of the finite-state continuous-time Markov process to obtain scheduling solutions for workflow tasks. It selects the lowest expenditure of the scheme of the resource system under the certain credit level of the reliability of resource combination on the critical path in DAG. The simulation shows the validity of the algorithm.

Key words: workflow scheduling, critical path, reliability of critical region

中图分类号: 

• TP393