混合型实时容错调度算法的设计和性能分析

以往文献中研究的实时容错调度算法都只能调度单一的具有容错需求的任务.该文建立了一个混合型实时容错调度模型,提出一种静态实时容错调度算法.该算法能同时调度具有容错需求的实时任务和无容错需求的实时任务.该文还提出了一个求解最小处理机个数的算法,用于对静态实时容错调度算法的性能进行模拟分析.为了提高静态调度算法的调度性能,提出了一种动态调度算法.最后，通过模拟实验分析了静态和动态调度算法的性能.实验表明,调度算法的性能与实时任务的个数、任务的计算时间、周期和处理机个数等系统参数相关.

Design and Performance Analysis of a Hybrid Real-Time Scheduling Algorithm with Fault-Tolerance

Since many real-time scheduling algorithms with fault-tolerance, reported in literature, can only schedule tasks with fault-tolerant requirements, the authors present a model of hybrid real-time fault-tolerant scheduling, and proposes a hybrid scheduling algorithm for real-time tasks in this paper. The static scheduling algorithm, a part of hybrid model can schedule tasks with fault-tolerant requirements together with those without fault-tolerant requirements. An algorithm, which is used to find out the minimal number of processors needed for the real-time tasks, is also presented in this paper, so the performance of the static scheduling algorithm can be simulated and analyzed. In order to enhance the performance of the static real-time scheduling algorithm with fault-tolerance, a dynamic scheduling algorithm is studied. The performance simulation and analysis of the scheduling algorithms are presented, and experiment results show that the performance is related with the number of tasks, computation time, period and the number of processors.