Search
Abstract
Imbalanced process arrival patterns (PAPs) are ubiquitous in many parallel and distributed systems, especially in HPC ones. The collective operations, e.g. in MPI, are designed for equal process arrival times (PATs), and are not optimized for deviations in their appearance. We propose eight new PAP-aware algorithms for the scatter and gather operations. They are binomial or linear tree adaptations introducing additional process ordering and (in some cases) additional activities in a special background thread. The solution was implemented using one of the most popular open source MPI compliant library (OpenMPI), and evaluated in a typical HPC environment using a specially developed benchmark as well as a real application: FFT. The experimental results show a significant advantage of the proposed approach over the default OpenMPI implementation, showing good scalability and high performance with the FFT acceleration for the communication run time: 16.7% and for the total application execution time: 3.3%.
Citations
-
2
CrossRef
-
0
Web of Science
-
2
Scopus
Cite as
Full text
- Publication version
- Accepted or Published Version
- License
-
open in new tab
Keywords
Details
- Category:
- Articles
- Type:
- artykuły w czasopismach
- Published in:
-
Cluster Computing-The Journal of Networks Software Tools and Applications
no. 23,
pages 2735 - 2751,
ISSN: 1386-7857 - Language:
- English
- Publication year:
- 2020
- Bibliographic description:
- Proficz J.: Process arrival pattern aware algorithms for acceleration of scatter and gather operations// Cluster Computing-The Journal of Networks Software Tools and Applications -Vol. 23, (2020), s.2735-2751
- DOI:
- Digital Object Identifier (open in new tab) 10.1007/s10586-019-03040-x
- Verified by:
- Gdańsk University of Technology
seen 13859 times