分布式实时系统容错算法的研究

发布时间：2018-06-18 07:26

本文选题：分布式实时系统 + 主副版本复制技术　；参考：《长春理工大学》2013年硕士论文

【摘要】：为了保障分布式实时系统的可靠性,本文采用主副版本复制技术来提高系统的容错能力。在此基础上,结合首次适应分配算法,提出了一种基于异构的分布式实时容错调度算法。综合考虑实时任务的时限、价值和松弛时间等特征因素,来计算任务的优先级,并将实时任务按照优先级进行排序。根据系统特点以及对实际应用的分析,给出系统容错模型,并以此给出任务容错可调度条件。在满足任务可调度的前提下,对主版本时限进行了优化,在一定程度上均衡处理器负载和提高处理器利用率。最后通过模拟实验,将本文算法与首次适应算法相比较。模拟实验表明,本文算法在任务接收率和容错性能上有所提升,可以得到比首次适应算法更高的处理器利用率。
[Abstract]:In order to ensure the reliability of distributed real-time system, this paper adopts master-subsidiary version replication technology to improve the fault-tolerant ability of the system. On the basis of this, a distributed real-time fault-tolerant scheduling algorithm based on heterogeneity is proposed in combination with the first-time adaptive allocation algorithm. Considering the time limit, value and relaxation time of the real-time task, the priority of the task is calculated, and the real-time task is sorted according to the priority. According to the characteristics of the system and the analysis of the practical application, the fault-tolerant model of the system is given, and the task fault-tolerant scheduling condition is given. On the premise that the task can be scheduled, the time limit of the main version is optimized to balance the processor load and improve the processor utilization to a certain extent. Finally, through the simulation experiment, the algorithm is compared with the first adaptive algorithm. The simulation results show that the proposed algorithm can improve the task acceptance rate and fault-tolerant performance and achieve higher processor utilization than the first-time adaptive algorithm.
【学位授予单位】：长春理工大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP302.8

【参考文献】