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

L. Chen, I. 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 Dec 1
Externally publishedYes
EventFourth Eurographics Workshop on Parallel Graphics and Visualization - Blaubeuren, Germany
Duration: 2002 Sep 92002 Sep 10

Publication series

NameEurographics Workshop on Parallel Graphics and Visualization

Other

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

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Parallel performance optimization of large-scale unstructured data visualization for the earth simulator'. Together they form a unique fingerprint.

Cite this