A compilation-based approach to performant reduction and redistribution collective communication algorithms
The International Journal of High Performance Computing Applications
Published online on August 01, 2025
Abstract
The International Journal of High Performance Computing Applications, Volume 40, Issue 2, Page 219-239, March 2026.
Modern supercomputers feature an ever-increasing degree of parallelism, particularly in the number of cores per node. These high core counts are considered in our flexible implementation of allreduce, which was implemented specifically with shared-memory ...
Modern supercomputers feature an ever-increasing degree of parallelism, particularly in the number of cores per node. These high core counts are considered in our flexible implementation of allreduce, which was implemented specifically with shared-memory ...