Leveraging MPI-3 Shared-Memory Extensions for Efficient PGAS Runtime Systems

The relaxed semantics and rich functionality of one-sided communication primitives of MPI-3 makes MPI an attractive candidate for the implementation of PGAS models. However, the performance of such implementation suffers from the fact, that current MPI RMA implementations typically have a large overhead when source and target of a communication request share a common, local physical memory. In this paper, we present an optimized PGAS-like runtime system which uses the new MPI-3 shared-memory extensions to serve intra-node communication requests and MPI-3 one-sided communication primitives to serve inter-node communication requests. The performance of our runtime system is evaluated on a Cray XC40 system through low-level communication benchmarks, a random-access benchmark and a stencil kernel. The results of the experiments demonstrate that the performance of our hybrid runtime system matches the performance of low-level RMA libraries for intra-node transfers, and that of MPI-3 for inter-node transfers.