Large-scale flow computation of complex geometries by building-cube method

Daisuke Sasaki, Shun Takahashi, Takashi Ishida, Kazuhiro Nakahashi, Hiroaki Kobayashi, Koki Okabe, Youichi Shimomura, Takashi Soga, Akihiko Musa

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

3 Citations (Scopus)

Abstract

Three-dimensional large scale incompressible flow simulation was conducted by Building-Cube Method (BCM) that is based on equally-spaced Cartesian mesh method. To exploit expected near-future high performance computer with massive number of processors, simple algorithms have been implemented in the BCM for mesh generation and flow solver. In this study, the capability of BCM for large-scale computation was demonstrated by solving formula 1 car model with around 200 million cells. The computation was conducted on Vector-Parallel Supercomputer system of NEC SX-9 at Cyberscience center of Tohoku University. The parallel efficiency of BCM with flat MPI and hybrid MPI on vector-parallel type system was also investigated.

Original languageEnglish
Title of host publicationHigh Performance Computing on Vector Systems 2009
PublisherSpringer Science and Business Media, LLC
Pages167-178
Number of pages12
ISBN (Print)9783642039126
DOIs
Publication statusPublished - 2010
Event2009 10th Teraflop Workshop - Stuttgart, Germany
Duration: 2009 Mar 162009 Mar 17

Publication series

NameHigh Performance Computing on Vector Systems 2009

Other

Other2009 10th Teraflop Workshop
Country/TerritoryGermany
CityStuttgart
Period09/3/1609/3/17

ASJC Scopus subject areas

  • Software
  • Mathematics(all)

Fingerprint

Dive into the research topics of 'Large-scale flow computation of complex geometries by building-cube method'. Together they form a unique fingerprint.

Cite this