固定优先级抢占调度算法下非周期任务实时性能研究

Research on Real-time Performance of Aperiodic Task for Fixed Priority Preemptive Scheduling Algorithm

嵌入式实时系统不仅要在功能上满足需求,而且要在性能上满足实时性需求.给定调度算法,实时性取决于各个任务的到达特征和执行时间.任务的到达特征由应用环境决定.为此,本文研究任务执行时间对实时性能的影响,为嵌入式系统设计提供参考.针对固定优先级抢占调度算法,应用排队论,提出一种非周期实时任务的理论模型.该理论模型包含两个优先级不同的非周期实时任务,给出了任务的执行时间长短对时限错过率、任务响应时间、任务队列长度等实时性能的影响.给出一个应用实例,仿真结果验证了理论模型的正确性.

Embedded real-time system should not only meet the functional requirements,but also meet the real-time requirements.Given the scheduling algorithm,the real-time performance depends on the arrival characteristics and execution time of each task.Application environment determines the characteristics of the task arrival.Therefore,this paper studies the effect of the task execution time on the real-time performance so as to provide some references for the embedded system design.For fixed priority preemptive scheduling algorithm,we apply queuing theory,and propose a theoretical model of aperiodic real-time tasks.The theoretical model contains two aperiodic real-time tasks whose priorities are different,and we give the effect of the task execution time length on real-time performance,which contains the deadline miss ratio,the task response time and the task queue length.This paper gives an application example,and the simulation results verify that the theoretical model is correct.