Parallel performance optimization of large-scale unstructured data visualization for the earth simulator

L. Chen, Issei Fujishiro, K. Nakajima

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

17 Citations (Scopus)

Abstract

This paper describes some efficient parallel performance optimization strategies for large-scale unstructured data visualization on SMP cluster machines including the Earth Simulator in Japan. The three-level hybrid parallelization is employed in our implementation, consisting of message passing for inter-SMP node communication, loop directives by OpenMP for intra-SMP node parallelization, and vectorization for each processing element (PE). In order to improve the speedup performance for the hybrid parallelization, some techniques, such as multi-coloring for removing data race and dynamic load repartition for load balancing, are considered. Good visualization images and high parallel performance have been achieved on Hitachi SR8000 for large-scale unstructured datasets, which shows the feasibility and effectiveness of our strategies.

Original languageEnglish
Title of host publicationEurographics Workshop on Parallel Graphics and Visualization
EditorsD. Bartz, X. Pueyo, E. Reinhard, S.N. Spencer
Pages133-140
Number of pages8
Publication statusPublished - 2002
Externally publishedYes
EventFourth Eurographics Workshop on Parallel Graphics and Visualization - Blaubeuren, Germany
Duration: 2002 Sep 92002 Sep 10

Other

OtherFourth Eurographics Workshop on Parallel Graphics and Visualization
CountryGermany
CityBlaubeuren
Period02/9/902/9/10

Fingerprint

Data visualization
Message passing
Coloring
Dynamic loads
Resource allocation
Visualization
Simulators
Earth (planet)
Communication
Processing

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Chen, L., Fujishiro, I., & Nakajima, K. (2002). Parallel performance optimization of large-scale unstructured data visualization for the earth simulator. In D. Bartz, X. Pueyo, E. Reinhard, & S. N. Spencer (Eds.), Eurographics Workshop on Parallel Graphics and Visualization (pp. 133-140)

Parallel performance optimization of large-scale unstructured data visualization for the earth simulator. / Chen, L.; Fujishiro, Issei; Nakajima, K.

Eurographics Workshop on Parallel Graphics and Visualization. ed. / D. Bartz; X. Pueyo; E. Reinhard; S.N. Spencer. 2002. p. 133-140.

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

Chen, L, Fujishiro, I & Nakajima, K 2002, Parallel performance optimization of large-scale unstructured data visualization for the earth simulator. in D Bartz, X Pueyo, E Reinhard & SN Spencer (eds), Eurographics Workshop on Parallel Graphics and Visualization. pp. 133-140, Fourth Eurographics Workshop on Parallel Graphics and Visualization, Blaubeuren, Germany, 02/9/9.
Chen L, Fujishiro I, Nakajima K. Parallel performance optimization of large-scale unstructured data visualization for the earth simulator. In Bartz D, Pueyo X, Reinhard E, Spencer SN, editors, Eurographics Workshop on Parallel Graphics and Visualization. 2002. p. 133-140
Chen, L. ; Fujishiro, Issei ; Nakajima, K. / Parallel performance optimization of large-scale unstructured data visualization for the earth simulator. Eurographics Workshop on Parallel Graphics and Visualization. editor / D. Bartz ; X. Pueyo ; E. Reinhard ; S.N. Spencer. 2002. pp. 133-140
@inproceedings{5609c2fe9c504958844680035fd8314a,
title = "Parallel performance optimization of large-scale unstructured data visualization for the earth simulator",
abstract = "This paper describes some efficient parallel performance optimization strategies for large-scale unstructured data visualization on SMP cluster machines including the Earth Simulator in Japan. The three-level hybrid parallelization is employed in our implementation, consisting of message passing for inter-SMP node communication, loop directives by OpenMP for intra-SMP node parallelization, and vectorization for each processing element (PE). In order to improve the speedup performance for the hybrid parallelization, some techniques, such as multi-coloring for removing data race and dynamic load repartition for load balancing, are considered. Good visualization images and high parallel performance have been achieved on Hitachi SR8000 for large-scale unstructured datasets, which shows the feasibility and effectiveness of our strategies.",
author = "L. Chen and Issei Fujishiro and K. Nakajima",
year = "2002",
language = "English",
isbn = "1581135793",
pages = "133--140",
editor = "D. Bartz and X. Pueyo and E. Reinhard and S.N. Spencer",
booktitle = "Eurographics Workshop on Parallel Graphics and Visualization",

}

TY - GEN

T1 - Parallel performance optimization of large-scale unstructured data visualization for the earth simulator

AU - Chen, L.

AU - Fujishiro, Issei

AU - Nakajima, K.

PY - 2002

Y1 - 2002

N2 - This paper describes some efficient parallel performance optimization strategies for large-scale unstructured data visualization on SMP cluster machines including the Earth Simulator in Japan. The three-level hybrid parallelization is employed in our implementation, consisting of message passing for inter-SMP node communication, loop directives by OpenMP for intra-SMP node parallelization, and vectorization for each processing element (PE). In order to improve the speedup performance for the hybrid parallelization, some techniques, such as multi-coloring for removing data race and dynamic load repartition for load balancing, are considered. Good visualization images and high parallel performance have been achieved on Hitachi SR8000 for large-scale unstructured datasets, which shows the feasibility and effectiveness of our strategies.

AB - This paper describes some efficient parallel performance optimization strategies for large-scale unstructured data visualization on SMP cluster machines including the Earth Simulator in Japan. The three-level hybrid parallelization is employed in our implementation, consisting of message passing for inter-SMP node communication, loop directives by OpenMP for intra-SMP node parallelization, and vectorization for each processing element (PE). In order to improve the speedup performance for the hybrid parallelization, some techniques, such as multi-coloring for removing data race and dynamic load repartition for load balancing, are considered. Good visualization images and high parallel performance have been achieved on Hitachi SR8000 for large-scale unstructured datasets, which shows the feasibility and effectiveness of our strategies.

UR - http://www.scopus.com/inward/record.url?scp=0036982662&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036982662&partnerID=8YFLogxK

M3 - Conference contribution

SN - 1581135793

SP - 133

EP - 140

BT - Eurographics Workshop on Parallel Graphics and Visualization

A2 - Bartz, D.

A2 - Pueyo, X.

A2 - Reinhard, E.

A2 - Spencer, S.N.

ER -