Treffer: Comparing Distributed Termination Detection Algorithms for Task-Based Runtime Systems on HPC platforms

Title:
Comparing Distributed Termination Detection Algorithms for Task-Based Runtime Systems on HPC platforms
Authors:
Bosilca, George, Bouteiller, Aurélien, Hérault, Thomas, Le Fèvre, Valentin, Robert, Yves, Dongarra, Jack, J
Contributors:
Innovative Computing Laboratory Knoxville (ICL), The University of Tennessee Knoxville, Barcelona Supercomputing Center - Centro Nacional de Supercomputacion (BSC-CNS), Optimisation des ressources : modèles, algorithmes et ordonnancement (ROMA), Laboratoire de l'Informatique du Parallélisme (LIP), École normale supérieure de Lyon (ENS de Lyon), Université de Lyon-Université de Lyon-Université Claude Bernard Lyon 1 (UCBL), Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure de Lyon (ENS de Lyon), Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Centre Inria de Lyon, Institut National de Recherche en Informatique et en Automatique (Inria), Université de Lyon-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), University of Manchester Manchester
Source:
ISSN: 2185-2839.
Publisher Information:
CCSD
Dept. of Computer Engineering, Hiroshima University
Publication Year:
2022
Collection:
HAL Lyon 1 (University Claude Bernard Lyon 1)
Subject Terms:
Termination detection credit distribution algorithms task-based HPC application control messages, Termination detection, credit distribution algorithms, task-based HPC application, control messages, [INFO]Computer Science [cs]
Document Type:
Fachzeitschrift article in journal/newspaper
Language:
English
Availability:
https://inria.hal.science/hal-03920388
https://inria.hal.science/hal-03920388v1/document
https://inria.hal.science/hal-03920388v1/file/ijnc22.pdf
Rights:
info:eu-repo/semantics/OpenAccess
Accession Number:
edsbas.4E90ADDF
Database:
BASE

Weitere Informationen

International audience ; This paper revisits distributed termination detection algorithms in the context of High-Performance Computing (HPC) applications. We introduce an efficient variant of the Credit Distribution Algorithm (CDA) and compare it to the original algorithm (HCDA) as well as to its two primary competitors: the Four Counters algorithm (4C) and the Efficient Delay-Optimal Distributed algorithm (EDOD). We analyze the behavior of each algorithm for some simplified task-based kernels and show the superiority of CDA in terms of the number of control messages. We then compare the implementation of these algorithms over a task-based runtime system, PaRSEC and show the advantages and limitations of each approach on a practical implementation.