摘要：

容错技术中硬件冗余会产生较高的设计和生产成本。针对该问题,提出一种改进的实时嵌入式系统容错优化方法,基于检查点容错技术综合分析系统故障性能、硬实时任务时间约束和软实时任务的效用函数值。以设计的容错模型为基础,计算系统故障概率保证其在故障最大概率值内,给出硬任务截止时间确定可调度性,并应用改进的禁忌搜索算法获得软任务效用函数最佳值,算法有2种简单的邻节点结构,其禁忌准则遵循邻节点方法禁忌,优化效率明显改善。实验结果表明,该方法可进行故障分析等综合分析,并能迅速获得最大效用函数值。

关键词: 容错优化方法, 实时嵌入式, 检查点技术, 系统故障分析, 实时调度算法

Abstract:

For the hardware redundancy fault tolerance technology tends to produce the defect of high design and high production costs,this paper puts forward an improved fault tolerance optimization method of real-time embedded system.The optimization method is based on checkpoint fault tolerance technology,and it comprehensively analyzes system failure performance,hard real-time task time constraints,and the utility function value of the soft real-time tasks.It is based on the design of fault tolerance model,the improved optimization method calculates the system failure probability,determines the schedulability of hard tasks,and uses an improved tabu search heuristic optimization to obtain utility function best value of soft tasks.The tabu search algorithm has two neighboring nodes of simple structure,it is tabu guidelines follow the tabu of adjacent node method and optimization efficiency is improved significantly.Experimental results show that the fault tolerance optimization method can carry on fault analysis and so on,and can quickly obtain maximum utility function.

Key words: fault tolerance optimization method, real-time embedded, checkpoint technology, system fault analysis, real-time scheduling algorithm

中图分类号: 

• TP302