The International Conference for High Performance Computing, Networking, Storage, and Analysis
The International Conference for High Performance Computing, Networking, Storage, and Analysis
Workshop: ExaMPI25: Workshop on Extreme Scale MPI
Authors: James B. White III (Oak Ridge National Laboratory (ORNL), National Center for Computational Sciences)
Abstract: Near the full scale of exascale supercomputers, latency can dominate the cost of all-to-all communication even for very large message sizes. We describe GPU-aware all-to-all implementations designed to reduce latency for large message sizes at extreme scales, and we present their performance using 65536 tasks (8192 nodes) on the Frontier supercomputer at the Oak Ridge Leadership Computing Facility. Two implementations perform best for different ranges of message size, and all outperform the vendor-provided MPI_Alltoall. Our results show promising options for improving implementations of MPI_Alltoall_init.
Back to ExaMPI25: Workshop on Extreme Scale MPI Archive Listing Back to Full Workshop Archive Listing