Towards a hybrid parallelization of lattice Boltzmann methods

Authors:
Vincent Heuveline;Mathias J. Krause;Jonas Latt
Affiliations:
Universität Karlsruhe (TH), Fritz-Erler-Str. 23, 76133 Karlsruhe, Germany;Universität Karlsruhe (TH), Fritz-Erler-Str. 23, 76133 Karlsruhe, Germany;Ecole Polytechnique Federale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland
Venue:
Computers & Mathematics with Applications
Year:
2009

Citing 5
Cited 4

An object oriented approach to lattice gas modeling

Future Generation Computer Systems - Special issue on high performance computing and networking Europe 1999
Parallel programming in OpenMP

Parallel programming in OpenMP
VLADYMIR: a C++ matrix library for data-parallel applications

Future Generation Computer Systems - Special issue: Computational science of lattice Boltzmann modelling
Performance Evaluation of Parallel Large-Scale Lattice Boltzmann Applications on Three Supercomputing Architectures

Proceedings of the 2004 ACM/IEEE conference on Supercomputing
Comparison of implementations of the lattice-Boltzmann method

Computers & Mathematics with Applications

Hybrid parallel simulations of fluid flows in complex geometries: application to the human lungs

Euro-Par 2010 Proceedings of the 2010 conference on Parallel processing
Optimized hybrid parallel lattice boltzmann fluid flow simulations on complex geometries

Euro-Par'12 Proceedings of the 18th international conference on Parallel Processing
Cellular Automata and GPGPU: An Application to Lava Flow Modeling

International Journal of Grid and High Performance Computing
Adjoint-based fluid flow control and optimisation with lattice Boltzmann methods

Computers & Mathematics with Applications

Quantified Score

Hi-index	0.09

Visualization

Abstract

Ongoing research towards the development of a hybrid parallelization concept for lattice Boltzmann methods is presented. It allows coping with platforms sharing both the properties of shared and distributed architectures. The proposed approach relies on spatial domain decomposition where each domain represents a basic block entity which is solved on a symmetric multi-processing (SMP) system. Emphasis is placed on the software design and the reworking needed to achieve good performance using OpenMP in that context. Those ideas are implemented in the C++ project OpenLB, which is also sketched in this article. The efficiency of the proposed approaches is tested on a 3D benchmark problem and compared with a purely MPI based approach.