Show simple item record

dc.contributor.authorChen, L.en_US
dc.contributor.authorFujishiro, I.en_US
dc.contributor.authorNakajima, K.en_US
dc.contributor.editorD. Bartz and X. Pueyo and E. Reinharden_US
dc.date.accessioned2014-01-26T16:21:13Z
dc.date.available2014-01-26T16:21:13Z
dc.date.issued2002en_US
dc.identifier.isbn1-58113-579-3en_US
dc.identifier.issn1727-348Xen_US
dc.identifier.urihttp://dx.doi.org/10.2312/EGPGV/EGPGV02/133-140en_US
dc.description.abstractThis 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.en_US
dc.publisherThe Eurographics Associationen_US
dc.titleParallel Performance Optimization of Large-Scale Unstructured Data Visualization for the Earth Simulatoren_US
dc.description.seriesinformationEurographics Workshop on Parallel Graphics and Visualizationen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record