This paper has considered the problem of preemptively scheduling a set of independent periodic hard real-time tasks in primary-backup based multiprocessor systems. An efficient scheduling algorithm-Task Partition based Fault Tolerant Rate-Monotonic (TPFTRM) is proposed which extends RM algorithm to primary-backup based multiprocessor to provide fault tolerance. Compared with previous scheduling algorithms in this area, TPFTRM abandons active backup copies and only uses passive and overlapping backup copies to maximize the backup over-booking and deallocation, thus improves the scheduling performance. Moreover, TPFTRM proposes the task partitioning and processors grouping technique, which reduce the scheduling computation time and also make an easy way to understand and implement it. Extensive simulations experiments are also carried out based on task sets with different parameters. And the simulation result shows a remarkable saving of processors as well as scheduling computation time compared with previous algorithms, which proves the feasibility and effectiveness of the proposed TPFTRM algorithm.%针对基于主副版本容错的多处理机中独立的、抢占性的硬实时任务,提出了一种高效的调度算法--TPFTRM(task partition based fault tolerant rate-monotonic)算法.该算法将单机实时RM算法扩展到容错多处理机上,并且调度过程中从不使用主动执行的任务副版本,而仅使用被动执行和主副重叠方式执行的任务副版本,从而最大限度地利用副版本重叠和分离技术提高了算法调度性能.此外,TPFTRM根据任务负载不同将任务集合划分成两个不相交的子集进行分配;还根据处理机调度的任务版本不同,将处理机集合划分成3个不相交的子集进行调度,从而使TPFTRM调度算法便于理解、实现以及减少了调度所需要的运行时间.模拟实验对各种具有不同周期和任务负载的任务集合进行了调度测试.实验结果表明,TPFTRM与目前所知同类算法相比,在调度相同参数的任务集合时不仅明显减少了调度所需要的处理机数目,还减少了调度所需要的运行时间,从而证实了TPFTRM算法的高效性.
展开▼
