基于关键路径和任务复制的多核调度算法

针对目前大多数多核处理器任务分配优化算法没有考虑关键路径上节点对任务完成时间的重要影响,导致任务完成总时间延迟的问题,提出了基于关键路径和任务复制(CPTD)的单任务调度算法。CPTD算法通过复制任务图中fork节点的方式将任务图转化为与之相对应的产品加工树;再在生成的产品加工树中找到关键路径,并采取使关键路径上节点的紧前节点尽早调度的方式,使关键路径上节点尽早开始执行,进而使产品加工树中节点完成时间得以提前,达到缩短任务执行总时间的目的。理论分析表明,CPTD算法能够实现应用程序在多核上充分并行处理,并能缩短任务完成时间。

A scheduling algorithm for multi-core based on critical path and task duplication

According to the problem of current scheduling algorithm for multi-core did not consider that the nodes on the critical path have a major impact on the ending time of tasks, led to task completion time delayed; a scheduling algorithm based on critical path and task duplication (CPTD) was proposed. Firstly duplicated fork-nodes to change the task graph into products processing tree, then found the critical path in the processing tree, and made the father nodes of the nodes on critical path began to work at the earliest time. These operations can advance the start time of nodes on critical path. The purpose of the above operation is to shorten the implementation of the mandate of the total time. Theoretical analysis shows that the algorithm can achieve a single task fully parallel processing on multi-core, also can shorten the completion time of the tasks.