• Revisiting credit distribution...

Treffer: Revisiting credit distribution algorithms for distributed termination detection

Title:
Revisiting credit distribution algorithms for distributed termination detection
Authors:
Bosilca, George, Bouteiller, Aurélien, Herault, Thomas, Fèvre, Valentin Le, Robert, Yves, Dongarra, Jack
Contributors:
Barcelona Supercomputing Center
Publisher Information:
Institute of Electrical and Electronics Engineers (IEEE)
Publication Year:
2021
Collection:
Universitat Politècnica de Catalunya, BarcelonaTech: UPCommons - Global access to UPC knowledge
Subject Terms:
Àrees temàtiques de la UPC::Informàtica::Arquitectura de computadors, Electronic data processing--Distributed processing, Heuristic algorithms, Parallel algorithms, High performance computing, Termination detection, Credit distribution algorithms, Task-based HPC application, Control messages, Supercomputadors
Document Type:
Konferenz conference object
File Description:
10 p.; application/pdf
Language:
English
Relation:
https://ieeexplore.ieee.org/document/9460615; http://hdl.handle.net/2117/358317
DOI:
10.1109/IPDPSW52791.2021.00095
Availability:
http://hdl.handle.net/2117/358317
https://doi.org/10.1109/IPDPSW52791.2021.00095
Rights:
Open Access
Accession Number:
edsbas.5CBB4C6A
Database:
BASE

Weitere Informationen

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. ; Peer Reviewed ; Postprint (author's final draft)