Tolerating Transient and Intermittent Failures

✍ Scribed by Sylvie Delaët; Sébastien Tixeuil

• Book ID: 102602468
• Publisher: Elsevier Science
• Year: 2002
• Tongue: English
• Weight: 284 KB
• Volume: 62
• Category: Article
• ISSN: 0743-7315
• DOI: 10.1006/jpdc.2001.1827

✦ Synopsis

Fault tolerance is a crucial property for recent distributed systems. We propose an algorithm that solves the census problem (list all processor identifiers and their relative distance) on an arbitrary strongly connected network.

This algorithm tolerates transient faults that corrupt the processors and communication links memory (it is self-stabilizing) as well as intermittent faults (fair loss, reorder, finite duplication of messages) on communication media. A formal proof establishes its correctness for the considered problem. Our algorithm leads to the construction of algorithms for any silent problems that are self-stabilizing while supporting the same communication hazards.