容错计算机系统中潜在故障的分析和处理途径

Analyzing and Dealing for Dormant Faults in a Fault-Tolerant Computer System

潜在故障是容错系统的潜在危害,因为大多数容错系统是基于单故障假设。以汽车导航系统为例来研究这一潜在危害,并用马尔可夫模型说明潜在故障恶化系统平均故障前时间。本文深入研究了一些可能的补救措施,其中透明的在线测试是最有效的方法之一,而用暂时离线的热贮备系统进行测试则是更可靠的方法。

Accumulation of dormant faults is a potential threat in a fault-tolerant system, because most fault tolerance systems are based on the single-fault assumption. This article investigates this threat with the example of an automotive steer-by-wire application based on the Time-Triggered Architecture and by means of a Markov model it illustrates that the effect of fault dormancy can degrade the MTTF of a system by several orders of magnitude. It studies some potential remedies, of which the transparent on-line testing is proved to be the most powerful one, while taking a hot spare system that offline temporarily to make the test is a more reliable way.