DOE OSTI · 1004675
MPI Support for Multi-core Architectures: Optimized Shared Memory Collectives
Abstract
With local core counts on the rise, taking advantage of shared memory to optimize collective operations can improve performance. We study several on-host shared memory optimized algorithms for MPI Bcast, MPI Reduce, and MPI Allreduce, using tree-based, and reduce-scatter algorithms. For small data operations with relatively large synchronization costs fan-in/fan-out algorithms generally perform best. For large messages data manipulation constitute the largest cost and reduce-scatter algorithms are best for reductions. These optimization improve performance by up to a factor of three. Memory and cache sharing effect require deliberate process layout and careful radix selection for tree-based methods
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Graham, Richard L, Shipman, Galen M. 2008-01-01. MPI Support for Multi-core Architectures: Optimized Shared Memory Collectives. https://searchnasa.com/record/osti%3A1004675/mpi-support-for-multi-core-architectures-optimized-shared-memory-collectives
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Search NASA by Indiciae: BibTeX · RIS.