新鲜度敏感的上下文推理实时调度算法
doi: 10.3724/SP.J.1146.2008.00510
Freshness-Aware Real-Time Scheduling Algorithm for Context Reasoning
-
摘要: 由于普适计算中上下文具有时效性,上下文推理器必须支持推理任务的实时调度。针对上下文推理结果在一段时间内仍然保持新鲜的特点,本文提出推理结果重复利用效率及其计算公式。在此基础上提出一种新鲜度敏感的上下文推理实时调度算法FRSA,以推理结果重复利用效率作为判断依据结合任务的deadline进行调度,其目标是在推理器负载较重时达到较高吞吐量。实验表明,在推理器负载重时,FRSA的系统吞吐量比经典调度算法(SJF,EDF,LSF和FCFS)高出10%-30%。Abstract: Due to the dynamic nature of contexts in pervasive computing, a context reasoner has to support real-time scheduling of reasoning jobs. Due to the fact that reasoning results remain fresh within a period of time, the concept of reasoning result reuse efficiency and its computation method are proposed. Then a Fresh-aware Real-time Scheduling Algorithm (FRSA) is proposed to promote the system throughput when the reasoner is overloaded, which schedules reasoning jobs according to their result reuse efficiencies and deadlines. The simulation demonstrates that when the reasoner is heavily overloaded, the throughput of FRSA is 10% to 30% better than those of classic scheduling algorithms SJF, EDF, LSF and FCFS.
-
李蕊, 李仁发. 上下文感知计算及系统框架综述. 计算机研究与发展, 2007, 44(2): 269-276.Li R and Li R F. A survey of context-aware computing and itssystem infrastructure. Journal of Computing Research andDevelopment, 2007, 44(2): 269-276.[2]Agostini A, Bettini C, and Riboni D. A performanceevaluation of ontology-based context reasoning. Proceedingsof the Fifth Annual IEEE International Conference onPervasive Computing and Communications Workshops(PerComW'07), New York,USA, 19 - 23 March 2007: 3-8.[3]Liu C L and Layland J. Scheduling algorithms formultiprogramming in real-time systems[J].Journal of the ACM.1973, 20(1):46-61[4]Dertouzos M L and Mok A K. Multiprocessor on-linescheduling of hard-real-time tasks. IEEE Trans. on SoftwareEngineering, 1989, 15(12): 1497-1506.[5]Silberschatz A.[J].Galvin P B, and Gagne G. Operating SystemConcepts, Sixth Edition. John Wiley Sons, Inc.2002,:-[6]Han Q and VenKatasubramanian N. Timeliness-accuracybalanced collection of dynamic context data[J].IEEE Trans. onParallel and Distributed Systems.2007, 18(2):158-171[7]Chen H L. An intelligent broker architecture for pervasivecontext-aware systems. [Ph.D. dissertation], Department ofComputer Science and Electrical Engineering, University ofMaryland, Baltimore County, 2004.
计量
- 文章访问数: 3139
- HTML全文浏览量: 78
- PDF下载量: 636
- 被引次数: 0