The Performance Value of Shared Network Caches in Clustered Multiprocessor Workstations

This paper evaluates the benefit of adding a shared cache to the network interface as a means of improving the performance of networked workstations configured as a distributed shared memory multiprocessor. A cache on the network interface, shared by all processors on each cluster, offers the potential benefits of retaining evicted processor cache lines, providing implicit prefetching when network cache lines are longer than processor cache lines, and increasing intra-cluster sharing. Using simulation, the performance of eight parallel scientific applications was evaluated. In each case, we examined in detail the means by which processor cache misses were satisfied. Our results were mixed. For the applications studied, we found that the network cache offers substantial performance benefit when processor caches are too small to hold the application's primary working set, or when network contention limits application performance. The expected benefits of implicit prefetching and increase..