实时任务优先级动态分配策略

Dynamic Priority Assignment Strategy for Real-time Task

现有大多数方法都是根据任务的截止期、空闲时间或者价值(密度)来确定任务的优先级,很少综合分析以上各个因素对任务优先级的影响.论文研究任务价值密度和紧迫性的变化特征,提出优先级的动态分派策略DPA,该策略可通过参数p与q来调节任务价值密度与紧迫性对任务优先级影响的程度,从而满足不同应用需求.最后,仿真实验结果显示,应用DPA策略的基于优先级的调度可以提高系统累积价值收益,降低任务截止期错失率.

Most of the existing methods assign the priorities of tasks according to the deadlines,slack times and values（density） of tasks,but seldom synthetically analyze the effects of the three factors on the priorities of tasks.This paper studies the dynamic value density and urgency of a task by studying its three attributes,proposes a dynamic priority assignment strategy DPA.In DPA strategy,two parameters p and q are used to adjust the weight that the value density and urgency of a task impact on its priority,by which,it is suitable for different applications.The experimental result shows that the performance on the accrual value and the deadline miss ratio of the system can be improved by applying DPA.