An effective parallel heuristic algorithm for the mapping problem

The mapping problem is a computationally intractable optimisation problem. Several heuristic methods aiming to obtain a `good´ suboptimal solution in a reasonable time have been proposed. We present a parallel heuristic optimisation technique for the mapping problem which is based on the principles of tabu searches. The adaptation of the tabu search strategy to this problem and its parallel approach based on locally linked searches are discussed. We empirically compare the performance of our strategy with that of a globally linked parallel strategy. Experimental findings are presented and analysed