基于RM策略的资源受限偶发任务调度算法  

Sporadic tasks scheduling algorithm with resources constraints based on RM scheme

作　　者：张忆文[1] 王成[1] 张惠臻[1] Zhang Yiwen Wang Cheng Zhang Huizhen(College of Computer Science and Technology, Huaqiao University, Xiamen, 361021, China)

机构地区：[1]华侨大学计算机科学与技术学院,福建厦门361021

出　　处：《华中科技大学学报（自然科学版）》2017年第7期115-121,共7页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(51305142;61502181);福建省科技计划引导性资助项目(2017H01010065);华侨大学引进人才科研启动金资助项目(16BS104)

摘　　要：针对现有资源受限偶发任务低能耗调度算法存在的不足,提出基于RM策略的资源受限偶发任务低能耗调度算法.首先,借鉴抢占阈值的思想,通过修改任务优先级的概念,提出双优先级单调速率调度(RM/DPP)算法,并且给出RM/DPP算法可行的必要条件.然后,以RM/DPP算法为基础,提出资源受限偶发任务低能耗调度(STLPSARC)算法,并且给出该算法可行的必要条件.最后,通过实验验证STLPSARC算法的性能,结果表明:STLPSARC算法比RM/DPP算法平均节约大约65.03%的能耗.Aiming at the shortcomings of sporadic task low energy scheduling algrithm with resoure- constration, a sporadic tasks low energy scheduling algorithm with resource-constrainted based on rate monotonic （RM） scheme was proposed . First, a double priority scheduling algorithm based on RM scheme （RM/DPP） was proposed, which modified the concept of the priority and used the idea of the preemption threshold. Moreover, the necessary condition of scheduling feasibility for the RM/DPP al- gorithm was given. Then, for energy efficiency, the sporadic tasks low energy scheduling algorithm with resources constraints （STLPSARC） based on the RM/DPP algorithm was proposed. Further- more, the necessary condition of scheduling feasibility for the STLPSARC algorithm was given. Fi- nally, an experiment was used to evaluate the performance of the STLPSARC algorithm. The experi- mental results show that the STLPSARC algorithm can reduce the energy consumption by 65.03% than the RM/DPP algorithm.

关 键 词：偶发任务 资源受限 能耗管理 实时调度 RM策略 

分 类 号：TP316.2[自动化与计算机技术—计算机软件与理论]