Parallel Particle Advection Bake-Off for Scientific Visualization Workloads

Roba Binyahib; David Pugmire; Abhishek Yenpure; Hank Childs

doi:10.1109/CLUSTER49012.2020.00048

Parallel Particle Advection Bake-Off for Scientific Visualization Workloads

Roba Binyahib, David Pugmire, Abhishek Yenpure, Hank Childs

Visualization

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

5 Scopus citations

Abstract

There are multiple algorithms for parallelizing particle advection for scientific visualization workloads. While many previous studies have contributed to the understanding of individual algorithms, our study aims to provide a holistic understanding of how algorithms perform relative to each other on various workloads. To accomplish this, we consider four popular parallelization algorithms and run a 'bake-off' study (i.e., an empirical study) to identify the best matches for each. The study includes 216 tests, going to a concurrency of up to 8192 cores and considering data sets as large as 34 billion cells with 300 million particles. Overall, our study informs three important research questions: (1) which parallelization algorithms perform best for a given workload?, (2) why?, and (3) what are the unsolved problems in parallel particle advection? In terms of findings, we find that the seeding box is the most important factor in choosing the best algorithm, and also that there is a significant opportunity for improvement in execution time, scalability, and efficiency.

Original language	English
Title of host publication	Proceedings - 2020 IEEE International Conference on Cluster Computing, CLUSTER 2020
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	381-391
Number of pages	11
ISBN (Electronic)	9781728166773
DOIs	https://doi.org/10.1109/CLUSTER49012.2020.00048
State	Published - Sep 2020
Event	22nd IEEE International Conference on Cluster Computing, CLUSTER 2020 - Kobe, Japan Duration: Sep 14 2020 → Sep 17 2020

Publication series

Name	Proceedings - IEEE International Conference on Cluster Computing, ICCC
Volume	2020-September
ISSN (Print)	1552-5244

Conference

Conference	22nd IEEE International Conference on Cluster Computing, CLUSTER 2020
Country/Territory	Japan
City	Kobe
Period	09/14/20 → 09/17/20

Funding

ACKNOWLEDGMENT This research was supported by the Exascale Computing Project (17-SC-20-SC), a collaborative effort of the U.S. Department of Energy Office of Science and the National Nuclear Security Administration. This research was supported by the Scientific Discovery through Advanced Computing (SciDAC) program of the U.S. Department of Energy. This research used resources of the National Energy Research Scientific Computing Center (NERSC), a U.S. Department of Energy Office of Science User Facility operated under Contract No. DE-AC02-05CH11231.

Keywords

Scientific visualization
flow visualization
parallel processing
particle advection

Access to Document

10.1109/CLUSTER49012.2020.00048

Cite this

Binyahib, R., Pugmire, D., Yenpure, A., & Childs, H. (2020). Parallel Particle Advection Bake-Off for Scientific Visualization Workloads. In Proceedings - 2020 IEEE International Conference on Cluster Computing, CLUSTER 2020 (pp. 381-391). Article 9229571 (Proceedings - IEEE International Conference on Cluster Computing, ICCC; Vol. 2020-September). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CLUSTER49012.2020.00048

Binyahib, Roba ; Pugmire, David ; Yenpure, Abhishek et al. / Parallel Particle Advection Bake-Off for Scientific Visualization Workloads. Proceedings - 2020 IEEE International Conference on Cluster Computing, CLUSTER 2020. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 381-391 (Proceedings - IEEE International Conference on Cluster Computing, ICCC).

@inproceedings{a292028325964133b7b1f725ab4fa0fc,

title = "Parallel Particle Advection Bake-Off for Scientific Visualization Workloads",

abstract = "There are multiple algorithms for parallelizing particle advection for scientific visualization workloads. While many previous studies have contributed to the understanding of individual algorithms, our study aims to provide a holistic understanding of how algorithms perform relative to each other on various workloads. To accomplish this, we consider four popular parallelization algorithms and run a 'bake-off' study (i.e., an empirical study) to identify the best matches for each. The study includes 216 tests, going to a concurrency of up to 8192 cores and considering data sets as large as 34 billion cells with 300 million particles. Overall, our study informs three important research questions: (1) which parallelization algorithms perform best for a given workload?, (2) why?, and (3) what are the unsolved problems in parallel particle advection? In terms of findings, we find that the seeding box is the most important factor in choosing the best algorithm, and also that there is a significant opportunity for improvement in execution time, scalability, and efficiency.",

keywords = "Scientific visualization, flow visualization, parallel processing, particle advection",

author = "Roba Binyahib and David Pugmire and Abhishek Yenpure and Hank Childs",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 22nd IEEE International Conference on Cluster Computing, CLUSTER 2020 ; Conference date: 14-09-2020 Through 17-09-2020",

year = "2020",

month = sep,

doi = "10.1109/CLUSTER49012.2020.00048",

language = "English",

series = "Proceedings - IEEE International Conference on Cluster Computing, ICCC",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "381--391",

booktitle = "Proceedings - 2020 IEEE International Conference on Cluster Computing, CLUSTER 2020",

}

Binyahib, R, Pugmire, D, Yenpure, A & Childs, H 2020, Parallel Particle Advection Bake-Off for Scientific Visualization Workloads. in Proceedings - 2020 IEEE International Conference on Cluster Computing, CLUSTER 2020., 9229571, Proceedings - IEEE International Conference on Cluster Computing, ICCC, vol. 2020-September, Institute of Electrical and Electronics Engineers Inc., pp. 381-391, 22nd IEEE International Conference on Cluster Computing, CLUSTER 2020, Kobe, Japan, 09/14/20. https://doi.org/10.1109/CLUSTER49012.2020.00048

Parallel Particle Advection Bake-Off for Scientific Visualization Workloads. / Binyahib, Roba; Pugmire, David; Yenpure, Abhishek et al.
Proceedings - 2020 IEEE International Conference on Cluster Computing, CLUSTER 2020. Institute of Electrical and Electronics Engineers Inc., 2020. p. 381-391 9229571 (Proceedings - IEEE International Conference on Cluster Computing, ICCC; Vol. 2020-September).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Parallel Particle Advection Bake-Off for Scientific Visualization Workloads

AU - Binyahib, Roba

AU - Pugmire, David

AU - Yenpure, Abhishek

AU - Childs, Hank

PY - 2020/9

Y1 - 2020/9

N2 - There are multiple algorithms for parallelizing particle advection for scientific visualization workloads. While many previous studies have contributed to the understanding of individual algorithms, our study aims to provide a holistic understanding of how algorithms perform relative to each other on various workloads. To accomplish this, we consider four popular parallelization algorithms and run a 'bake-off' study (i.e., an empirical study) to identify the best matches for each. The study includes 216 tests, going to a concurrency of up to 8192 cores and considering data sets as large as 34 billion cells with 300 million particles. Overall, our study informs three important research questions: (1) which parallelization algorithms perform best for a given workload?, (2) why?, and (3) what are the unsolved problems in parallel particle advection? In terms of findings, we find that the seeding box is the most important factor in choosing the best algorithm, and also that there is a significant opportunity for improvement in execution time, scalability, and efficiency.

AB - There are multiple algorithms for parallelizing particle advection for scientific visualization workloads. While many previous studies have contributed to the understanding of individual algorithms, our study aims to provide a holistic understanding of how algorithms perform relative to each other on various workloads. To accomplish this, we consider four popular parallelization algorithms and run a 'bake-off' study (i.e., an empirical study) to identify the best matches for each. The study includes 216 tests, going to a concurrency of up to 8192 cores and considering data sets as large as 34 billion cells with 300 million particles. Overall, our study informs three important research questions: (1) which parallelization algorithms perform best for a given workload?, (2) why?, and (3) what are the unsolved problems in parallel particle advection? In terms of findings, we find that the seeding box is the most important factor in choosing the best algorithm, and also that there is a significant opportunity for improvement in execution time, scalability, and efficiency.

KW - Scientific visualization

KW - flow visualization

KW - parallel processing

KW - particle advection

UR - http://www.scopus.com/inward/record.url?scp=85096221686&partnerID=8YFLogxK

U2 - 10.1109/CLUSTER49012.2020.00048

DO - 10.1109/CLUSTER49012.2020.00048

M3 - Conference contribution

AN - SCOPUS:85096221686

T3 - Proceedings - IEEE International Conference on Cluster Computing, ICCC

SP - 381

EP - 391

BT - Proceedings - 2020 IEEE International Conference on Cluster Computing, CLUSTER 2020

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 22nd IEEE International Conference on Cluster Computing, CLUSTER 2020

Y2 - 14 September 2020 through 17 September 2020

ER -

Binyahib R, Pugmire D, Yenpure A, Childs H. Parallel Particle Advection Bake-Off for Scientific Visualization Workloads. In Proceedings - 2020 IEEE International Conference on Cluster Computing, CLUSTER 2020. Institute of Electrical and Electronics Engineers Inc. 2020. p. 381-391. 9229571. (Proceedings - IEEE International Conference on Cluster Computing, ICCC). doi: 10.1109/CLUSTER49012.2020.00048

Parallel Particle Advection Bake-Off for Scientific Visualization Workloads

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this