首页> 外文会议>Fault-Tolerant Computing, 1994. FTCS-24. Digest of Papers., Twenty-Fourth International Symposium on >Analysis and experimental evaluation of comparison-basedsystem-level diagnosis for multiprocessor systems
【24h】

Analysis and experimental evaluation of comparison-basedsystem-level diagnosis for multiprocessor systems

机译:基于比较的分析和实验评估多处理器系统的系统级诊断

获取原文

摘要

A comparison-based model for system-level fault diagnosis thatgeneralizes both the classical PMC model and the Maeng/Malek comparisonmodel is studied. A new necessary and sufficient condition for a systemto be t-diagnosable under this model is proven. Also, a class of systemsthat uses the minimum number of communication links to obtain a givendegree of diagnosability is presented. Next, a distributed diagnosisalgorithm is presented that can reduce the number of tests necessary fordiagnosis when the number of faults is relatively small. To demonstratethe practicality of our diagnosis approach, a fault table baseddiagnosis algorithm suitable for relatively small systems has beenimplemented in the Common Spaceborne Multicomputer Operating System(COSMOS). A simulator for the JPL MAX multicomputer system runningCOSMOS was used to test the algorithm and evaluate its performance. Theresults show that the algorithm diagnoses all fault situations with lowlatency and very little overhead
机译:基于比较的系统级故障诊断模型 概括了经典PMC模型和Maeng / Malek比较 模型进行了研究。系统的新的必要条件和充分条件 在这种模式下可以被t诊断。还有一类系统 使用最少数量的通信链接获得给定的 介绍了可诊断性程度。接下来,进行分布式诊断 提出了一种可以减少必要的测试次数的算法。 诊断时的故障数量比较少。展示 诊断方法的实用性,基于故障表 适用于相对较小系统的诊断算法已经 在通用星空多计算机操作系统中实现 (宇宙)。运行JPL MAX多计算机系统的模拟器 COSMOS用于测试算法并评估其性能。这 结果表明,该算法可以对所有故障情况进行低诊断 延迟和很少的开销

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号