Performance Evaluation of Parallel Large-Scale Lattice Boltzmann Applications on Three Supercomputing Architectures

  • Authors:

  • Thomas Pohl;Frank Deserno;Nils Thurey;Ulrich Rude;Peter Lammers;Gerhard Wellein;Thomas Zeiser

  • Affiliations:

  • University of Erlangen;University of Erlangen;University of Erlangen;University of Erlangen;Höchstleistungsrechenzentrum Stuttgart;Regionales Rechenzentrum Erlangen;Regionales Rechenzentrum Erlangen

  • Venue:
  • Proceedings of the 2004 ACM/IEEE conference on Supercomputing
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Computationally intensive programs with moderate communication requirements such as CFD codes suffer from the standard slow interconnects of commodity "off the shelf" (COTS) hardware. We will introduce different large-scale applications of the Lattice Boltzmann Method (LBM) in fluid dynamics, material science, and chemical engineering and present results of the parallel performance on different architectures. It will be shown that a high speed communication network in combination with an efficient CPU is mandatory in order to achieve the required performance. An estimation of the necessary CPU count to meet the performance of 1 TFlop/s will be given as well as a prediction as to which architecture is the most suitable for LBM. Finally, ratios of costs to application performance for tailored HPC systems and COTS architectures will be presented.