Project Description
MPI is the de facto standard programming model for large-scale scientific computing today. While we believe MPI is a viable programming model at exascale, both the MPI standard and MPICH have to address the challenges posed by of applications targeted at exascale systems. Research in this project tackles various challenges related to the efficiency of MPI on next-generation HPC systems.
Testbed
Early access systems (Iris) and all systems with recent network fabric and decent core counts (Skylake, KNL, it, Thing, Gomez, etc.)