Adaptive MPI Multirail Tuning for Non-uniform Input/Output Access

Multicore processors have not only reintroduced Non-Uniform Memory Access (NUMA) architectures in nowadays parallel computers, but they are also responsible for non-uniform access times with respect to Input/Output devices (NUIOA). In clusters of multicore machines equipped with several Network Interfaces, performance of communication between processes thus depends on which cores these processes are scheduled on, and on their distance to the Network Interface Cards involved. We propose a technique allowing multirail communication between processes to carefully distribute data among the network interfaces so as to counterbalance NUIOA effects. We demonstrate the relevance of our approach by evaluating its implementation within OpenMPI on a Myri-10G + InfiniBand cluster.