Hunold, S. (2023). Verifying Performance Guidelines for MPI Collectives at Scale. In Proceedings of 2023 SC23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis (SC23 Workshops) (pp. 1264–1268). ACM. https://doi.org/10.1145/3624062.3625532
MPI collective communication operations are crucial for high-performance computing, making the efficient implementation of collective algorithms essential for optimal application performance. While most MPI libraries provide several algorithms for a specific collective operation, each may work better in a specific scenario. Therefore, selecting the most suitable algorithm for each use case is important. However, even the best algorithm in a given MPI library’s set may deliver suboptimal performance.
Self-consistent MPI performance guidelines are general expectations that collectives must meet to be deemed performance-consistent. Specifically, a specialized collective call should not be slower than its less specialized counterparts. This paper introduces a tool for assessing the performance consistency of MPI collectives in a statistically sound manner. Through a case study, we demonstrate the current state of MPI performance consistency for three TOP500 machines.
en
Project title:
Offline- und Online-Autotuning von Parallelen Programmen: P33884-N (FWF - Österr. Wissenschaftsfonds)
CC BY 4.0
