Analyzing the parallel scalability of an implicit unstructured mesh CFD code

W. D. Gropp, D. K. Kaushik, D. E. Keyes, B. F. Smith

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

In this paper, we identify the scalability bottlenecks of an unstructured grid CFD code (PETSc-FUN3D) by studying the impact of several algorithmic and architectural parameters and by examiningdif ferent programmingmodels. We discuss the basic performance characteristics of this PDE code with the help of simple performance models developed in our earlier work, presentingprimarily experimental results. In addition to achievingg ood per-processor performance (which has been addressed in our cited work and without which scalability claims are suspect) we strive to improve the implementation and convergence scalability of PETSc-FUN3D on thousands of processors.

Original languageEnglish
Title of host publicationHigh Performance Computing - HiPC 2000 - 7th International Conference, Proceedings
PublisherSpringer Verlag
Pages395-404
Number of pages10
Volume1970
ISBN (Print)3540414290, 9783540414292
Publication statusPublished - 2000
Externally publishedYes
Event7th International Conference on High Performance Computing, HiPC 2000 - Bangalore, India
Duration: 17 Dec 200020 Dec 2000

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1970
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other7th International Conference on High Performance Computing, HiPC 2000
CountryIndia
CityBangalore
Period17/12/0020/12/00

Fingerprint

Unstructured Mesh
Scalability
Computational fluid dynamics
Unstructured Grid
Performance Model
Experimental Results

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Gropp, W. D., Kaushik, D. K., Keyes, D. E., & Smith, B. F. (2000). Analyzing the parallel scalability of an implicit unstructured mesh CFD code. In High Performance Computing - HiPC 2000 - 7th International Conference, Proceedings (Vol. 1970, pp. 395-404). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1970). Springer Verlag.

Analyzing the parallel scalability of an implicit unstructured mesh CFD code. / Gropp, W. D.; Kaushik, D. K.; Keyes, D. E.; Smith, B. F.

High Performance Computing - HiPC 2000 - 7th International Conference, Proceedings. Vol. 1970 Springer Verlag, 2000. p. 395-404 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1970).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Gropp, WD, Kaushik, DK, Keyes, DE & Smith, BF 2000, Analyzing the parallel scalability of an implicit unstructured mesh CFD code. in High Performance Computing - HiPC 2000 - 7th International Conference, Proceedings. vol. 1970, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1970, Springer Verlag, pp. 395-404, 7th International Conference on High Performance Computing, HiPC 2000, Bangalore, India, 17/12/00.
Gropp WD, Kaushik DK, Keyes DE, Smith BF. Analyzing the parallel scalability of an implicit unstructured mesh CFD code. In High Performance Computing - HiPC 2000 - 7th International Conference, Proceedings. Vol. 1970. Springer Verlag. 2000. p. 395-404. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Gropp, W. D. ; Kaushik, D. K. ; Keyes, D. E. ; Smith, B. F. / Analyzing the parallel scalability of an implicit unstructured mesh CFD code. High Performance Computing - HiPC 2000 - 7th International Conference, Proceedings. Vol. 1970 Springer Verlag, 2000. pp. 395-404 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{911bf16702414a5f998b6384bb137f15,
title = "Analyzing the parallel scalability of an implicit unstructured mesh CFD code",
abstract = "In this paper, we identify the scalability bottlenecks of an unstructured grid CFD code (PETSc-FUN3D) by studying the impact of several algorithmic and architectural parameters and by examiningdif ferent programmingmodels. We discuss the basic performance characteristics of this PDE code with the help of simple performance models developed in our earlier work, presentingprimarily experimental results. In addition to achievingg ood per-processor performance (which has been addressed in our cited work and without which scalability claims are suspect) we strive to improve the implementation and convergence scalability of PETSc-FUN3D on thousands of processors.",
author = "Gropp, {W. D.} and Kaushik, {D. K.} and Keyes, {D. E.} and Smith, {B. F.}",
year = "2000",
language = "English",
isbn = "3540414290",
volume = "1970",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "395--404",
booktitle = "High Performance Computing - HiPC 2000 - 7th International Conference, Proceedings",

}

TY - GEN

T1 - Analyzing the parallel scalability of an implicit unstructured mesh CFD code

AU - Gropp, W. D.

AU - Kaushik, D. K.

AU - Keyes, D. E.

AU - Smith, B. F.

PY - 2000

Y1 - 2000

N2 - In this paper, we identify the scalability bottlenecks of an unstructured grid CFD code (PETSc-FUN3D) by studying the impact of several algorithmic and architectural parameters and by examiningdif ferent programmingmodels. We discuss the basic performance characteristics of this PDE code with the help of simple performance models developed in our earlier work, presentingprimarily experimental results. In addition to achievingg ood per-processor performance (which has been addressed in our cited work and without which scalability claims are suspect) we strive to improve the implementation and convergence scalability of PETSc-FUN3D on thousands of processors.

AB - In this paper, we identify the scalability bottlenecks of an unstructured grid CFD code (PETSc-FUN3D) by studying the impact of several algorithmic and architectural parameters and by examiningdif ferent programmingmodels. We discuss the basic performance characteristics of this PDE code with the help of simple performance models developed in our earlier work, presentingprimarily experimental results. In addition to achievingg ood per-processor performance (which has been addressed in our cited work and without which scalability claims are suspect) we strive to improve the implementation and convergence scalability of PETSc-FUN3D on thousands of processors.

UR - http://www.scopus.com/inward/record.url?scp=84947572059&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947572059&partnerID=8YFLogxK

M3 - Conference contribution

SN - 3540414290

SN - 9783540414292

VL - 1970

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 395

EP - 404

BT - High Performance Computing - HiPC 2000 - 7th International Conference, Proceedings

PB - Springer Verlag

ER -