Habich J, Zeiser T, Hager G, Wellein G (2011)
Publication Language: English
Publication Type: Conference contribution
Publication year: 2011
Publisher: Elsevier
Edited Volumes: Advances in Engineering Software
City/Town: ScienceDirect
Book Volume: 42
Pages Range: 266-272
Edition: 5
Conference Proceedings Title: Advances in Engineering Software
URI: http://www.sciencedirect.com/science/article/pii/S0965997810001274
DOI: 10.1016/j.advengsoft.2010.10.007
This paper presents implementation strategies and optimization approaches for a D3Q19 lattice Boltzmann flow solver on nVIDIA graphics processing units (GPUs). Using the STREAM benchmarks we demonstrate the GPU parallelization approach and obtain an upper limit for the flow solver performance. We discuss the GPU-specific implementation of the solver with a focus on memory alignment and register shortage. The optimized code is up to an order of magnitude faster than standard two-socket x86 servers with AMD Barcelona or Intel Nehalem CPUs. We further analyze data transfer rates for the PCI-express bus to evaluate the potential benefits of multi-GPU parallelism in a cluster environment. © 2010 Civil-Comp Ltd and Elsevier Ltd. All rights reserved.
APA:
Habich, J., Zeiser, T., Hager, G., & Wellein, G. (2011). Performance analysis and optimization strategies for a D3Q19 lattice Boltzmann kernel on nVIDIA GPUs using CUDA. In Advances in Engineering Software (pp. 266-272). ScienceDirect: Elsevier.
MLA:
Habich, Johannes, et al. "Performance analysis and optimization strategies for a D3Q19 lattice Boltzmann kernel on nVIDIA GPUs using CUDA." Proceedings of the PARENG 2009 ScienceDirect: Elsevier, 2011. 266-272.
BibTeX: Download