A new approach to the lattice Boltzmann method for graphics processing units

Christian Obrecht 1 Frédéric Kuznik 1 Bernard Tourancheau 2, 3, 4 Jean-Jacques Roux 1
2 SWING - Smart Wireless Networking
Inria Grenoble - Rhône-Alpes, CITI - CITI Centre of Innovation in Telecommunications and Integration of services
4 GRAAL - Algorithms and Scheduling for Distributed Heterogeneous Platforms
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
Abstract : Emerging many-core processors, like CUDA capable nVidia GPUs, are promising platforms for regular parallel algorithms such as the Lattice Boltzmann Method (LBM). Since the global memory for graphic devices shows high latency and LBM is data intensive, the memory access pattern is an important issue for achieving good performances. Whenever possible, global memory loads and stores should be coalescent and aligned, but the propagation phase in LBM can lead to frequent misaligned memory accesses. Most previous CUDA implementations of 3D LBM addressed this problem by using low latency on chip shared memory. Instead of this, our CUDA implementation of LBM follows carefully chosen data transfer schemes in global memory. For the 3D lid-driven cavity test case, we obtained up to 86% of the global memory maximal throughput on nVidia's GT200. We show that as a consequence highly efficient implementations of LBM on GPUs are possible, even for complex models.
Complete list of metadatas

Cited literature [17 references]  Display  Hide  Download

https://hal.inria.fr/inria-00568674
Contributor : Bernard Tourancheau <>
Submitted on : Tuesday, March 1, 2011 - 8:39:06 AM
Last modification on : Monday, December 10, 2018 - 10:54:03 AM
Long-term archiving on : Monday, May 30, 2011 - 2:22:49 AM

File

obrecht10a-HAL.pdf
Files produced by the author(s)

Identifiers

Citation

Christian Obrecht, Frédéric Kuznik, Bernard Tourancheau, Jean-Jacques Roux. A new approach to the lattice Boltzmann method for graphics processing units. Computers and Mathematics with Applications, Elsevier, 2010, ⟨10.1016/j.camwa.2010.01.054⟩. ⟨inria-00568674⟩

Share

Metrics

Record views

566

Files downloads

799