|
|
Numerical methods and programming, 2016, Volume 17, Issue 3, Pages 329–338
(Mi vmp840)
|
|
|
|
Efficiency analysis of hydrodynamic calculations on GPU and CPU clusters
A. V. Sentyabova, A. A. Gavrilova, M. A. Krivovb, A. A. Dektereva, M. N. Pritulab a S.S. Kutateladze Institute of Thermophysics, Siberian Division of the Russian Academy of Sciences
b TTG Labs Limited Liability Company, Moscow
Abstract:
Speedup of parallel hydrodynamic calculations on clusters with CPUs and GPUs is considered. The CFD SigmaFlow code developed by the authors and ported for GPU by means of CUDA is used in test calculations. The incompressible flow simulation is based on a SIMPLE-like procedure and on a discretization by the control volume method on unstructured hexahedral meshes. The performance evaluation shows a high efficiency of the new generation of GPUs for GPGPU calculations.
Keywords:
GPGPU, numerical simulation, computational fluid dynamics, SIMPLE, MPI, CUDA.
Received: 22.07.2016
Citation:
A. V. Sentyabov, A. A. Gavrilov, M. A. Krivov, A. A. Dekterev, M. N. Pritula, “Efficiency analysis of hydrodynamic calculations on GPU and CPU clusters”, Num. Meth. Prog., 17:3 (2016), 329–338
Linking options:
https://www.mathnet.ru/eng/vmp840 https://www.mathnet.ru/eng/vmp/v17/i3/p329
|
| Statistics & downloads: |
| Abstract page: | 208 | | Full-text PDF : | 92 | | References: | 3 |
|