Automatic Discovery of Collective Communication Patterns in Parallelized Task Graphs
International Journal of Parallel Programming · 2024-06 · 52 (3) · 171-186
DOI: 10.1007/s10766-024-00767-y ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Parallel Programming, online erschienen am 22.03.2024. Verfasst von Fabian Knorr und 3 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 14 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European High-Performance Computing Joint Undertaking, University of Innsbruck and Medical University of Innsbruck.
Abstract (Originalsprache)
Collective communication APIs equip MPI vendors with the necessary context to optimize cluster-wide operations on the basis of theoretical complexity models and characteristics of the involved interconnects. Modern HPC runtime systems with a programmability focus can perform dependency analysis to eliminate the need for manual communication entirely. Profiting from optimized collective routines in this context often requires global analysis of the implicit point-to-point communication pattern or tight constrains on the data access patterns allowed inside kernels. The Celerity API provides a high degree of freedom for both runtime implementors and application developers by tieing transparent work assignment to data access patterns through user-defined range-mapper functions. Canonically, data dependencies are resolved through an intra-node coherence model and inter-node point-to-point communication. This paper presents Collective Pattern Discovery (CPD), a fully distributed, coordination-free method for detecting collective communication patterns on parallelized task graphs. Through extensive scheduling and communication microbenchmarks as well as a strong scaling experiment on a compute-intensive application, we demonstrate that CPD can achieve substantial performance gains in the Celerity model.
Förderung
- European High-Performance Computing Joint Undertaking
- University of Innsbruck and Medical University of Innsbruck
Zitieren
APA
Knorr, F., Salzmann, P., Thoman, P., & Fahringer, T. (2024). Automatic Discovery of Collective Communication Patterns in Parallelized Task Graphs. International Journal of Parallel Programming, 52(3), 171-186. https://doi.org/10.1007/s10766-024-00767-y
Vancouver
Knorr F, Salzmann P, Thoman P, Fahringer T. Automatic Discovery of Collective Communication Patterns in Parallelized Task Graphs. International Journal of Parallel Programming. 2024;52(3):171-186. doi:10.1007/s10766-024-00767-y
BibTeX
@article{Knorr2024,
author = {Knorr, Fabian and Salzmann, Philip and Thoman, Peter and Fahringer, Thomas},
title = {Automatic Discovery of Collective Communication Patterns in Parallelized Task Graphs},
journal = {International Journal of Parallel Programming},
year = {2024},
volume = {52},
number = {3},
pages = {171--186},
doi = {10.1007/s10766-024-00767-y}
}
EUVIMED