AUV协同设计平台中多任务流调度算法研究

Research on multiple task flows scheduling algorithm in collaborative design platform for AUV

对AUV协同设计平台中多个任务流的调度问题进行建模,将其转换为分布式计算环境下的独立任务在线调度问题。针对系统异构和任务流具有优先级属性的特殊性,提出了一种基于预测的多任务流调度算法,采用统计和预测的方法评估各工作站执行任务的效用,并设计优先级策略和暂停调度策略,保证具有较高优先级的任务流较早分配和执行。实验结果表明,该算法在参数选取适当的情况下,性能优于传统的MCT和MET任务调度算法。

This paper modeled the multiple task flows scheduling problem in the collaborative design platform for AUV and transformed it into the independent tasks online scheduling problem in the distributed computing environment. Aiming at the particularity including the system was heterogeneous and the task flow had priority attribute, and proposed a prediction-based multiple task flows scheduling algorithm. Evaluated the utility of executing the task on each workstation using statistics and prediction method to improve the performance of task scheduling and designed the priority strategy to ensure the task flow with a higher priority could be assigned and executed earlier. The experimental results show that this algorithm performs better than some classical algorithms such as MCT and MET task scheduling algorithms when the parameters are well selected.