A Genetic-Based Fault-Tolerant Routing Strategy for Multiprocessor Networks

Abstract. We have investigated the adaptation of AI-based search techniques as topologyindependent fault-tolerant routing strategies on multiprocessor networks [9]. The results showed that these search techniques are suitable for adaptation, as fault-tolerant routing strategies with the exception that the routes obtained were non-minimal. In this research, we investigate the adaptation of a genetic-heuristic algorithm combination as a fault-tolerant routing strategy. Our results show that such a hybrid strategy results in a viable fault-tolerant routing strategy, which produces minimal or near-minimal routes with a corresponding significant reduction in the number of redundant node traversals. Under certain fault conditions, this new hybrid routing strategy outperforms the purely heuristic ones.

• Type: Conference
• Year: 1999
• Where: IPPS
• Authors: Peter K. K. Loh, Venson Shaw
• Keywords: Distributed And Parallel Computing, Fault-tolerant Routing, Fault-tolerant Routing Strategies, Fault-tolerant Routing Strategy, IPPS 1999