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.