MMLRU selection function: An output selection function on adaptive routing

Recently, adaptive routing algorithms which achieve a high performance compared with deterministic ones are commonly used in large scale parallel machines. Although their performance is influenced with an output selection function (output channel selection algorithm), only a few studies depending on specific topologies or specific traffic patterns have been done. In this paper, we propose an output selection function called MMLRU(Minimal Multiplexed and Least Recently Used) selection function, which is not based on a specific topology nor traffic pattern. Result of simulations shows that proposed output selection function achieves better performance than traditional ones with each traffic pattern and topology.