Abstract
The use of general purpose graphics processors for the numerical solution of problems in dynamics of viscous incompressible fluid is discussed. Specific features of the parallel implementation of the splitting scheme (projection method) are considered. The system of difference equations produced by the discretization of Poisson’s equations for pressure is solved using the multigrid method. A number of benchmark problems are solved on graphics processors, and approaches to the optimization of program code by using different memory types are discussed. The speedup of computations on graphics processors is compared with the computations on the central processor using grids with different resolution and different decompositions of the initial data into blocks.
Similar content being viewed by others
REFERENCES
J. D. Owens, D. Luebke, N. Govindaraju, M. Harris, J. Kruger, A. E. Lefohn, and T. J. Purcel, “A survey of general-purpose computation on graphics hardware,” Comput. Graphics Forum 26, 80–113 (2007).
A. V. Boreskov and A. Kharlamov, Basics of CUDA Technology (DMK, Moscow, 2010) [in Russian].
J. Sanders and E. Kandrot, CUDA by Example: An Introduction to General-Purpose GPU Programming (Addison-Wesley, Upper Saddle River, 2010).
K. N. Volkov, Yu. N. Deryugin, V. N. Emel’yanov, A. G. Karpenko, A. S. Kozelkov, and I. V. Teterina, Acceleration of Fluid Dynamics Computations on Unstructured Grids (Fizmatlit, Moscow, 2014).
A. V. Gorobets, S. A. Sukov, A. O. Zheleznyakov, P. B. Bogdanov, and B. N. Chetverushkin, “Using GPUs for hybrid two-level parallelization MPI+OpenMP on heterogeneous computer systems,” in Parallel Computational Technologies (Yuzhnoural’skii Gos Univ., Chelyabinsk, 2011), pp. 452–460.
K. N. Volkov, V. N. Emel’yanov, A. G. Karpenko, P. G. Smirnov, and I. V. Teterina, “Imlementation of the finite volume method and computation of flows of viscous compressible fluid on graphics processors,” Vychisl Meth. Program. 14 (1), 183–194 (2013).
A. Corrigan, F. Camelli, R. Lőhner, and J. Wallin, “Running unstructured grid-based CFD solvers on modern graphics hardware,” AIAA Paper No. 2009-4001 (2009).
I. C. Kampolis, X. S. Trompoukis, V. G. Asouti, and K. C. Giannakoglou, “D-based analysis and two-level aerodynamic optimization on graphics processing units,” Comput. Meth. Appl, Mech. Eng. 199, 712–722 (2010).
L. Fu, Z. Gao, K. Xu, and F. Xu, “A multi-block viscous flow solver based on GPU parallel methodology,” Comput. Fluids 95, 19–39 (2014).
M. Krotkiewski and M. Dabrowski, “Efficient 3D stencil computations using CUDA,” Parallel Comput. 39, 533–548 (2013).
J. Meng and K. Skadron, “A performance study for iterative stencil loops on GPUs with ghost zone optimizations,” Int. J. Parallel Program. 39 (1), 115–142 (2011).
D. A. Jacobsen and I. Senocak, “Multi-level parallelism for incompressible flow computations on GPU clusters,” Parallel Comput. 39, 1–20 (2013).
M. Tuttafesta, G. Colonna, and G. Pascazio, “Computing unsteady compressible flows using Roe’s flux-difference splitting scheme on GPUs,” Comput. Phys. Commun. 194, 1497–1510 (2013).
A. J. Chorin, “Numerical solution of Navier–Stokes equations,” Math. Comput. 22, 745–762 (1968).
O. M. Belotserkovskii, Numerical Simulation in Fluid Continuum Mechanics (Fizmatlit, Moscow, 1994) [in Russian].
K. N. Volkov, “An implementation of the splitting scheme on a staggered grid for computing unsteady flows of viscous incompressible fluid,” Vychisl. Metody Program. 6 (1), 269–282 (2005).
J. C. Thibault and I. Senocak, “CUDA implementation of a Navier–Stokes solver on multi-GPU desktop platforms for incompressible flows,” AIAA Paper., No. 2009-758 (2009).
K. N. Volkov and V. N. Emel’yanov, Simulation of Large-Scale Vortices in the Computations of Turbulent Flows (Fizmatlit, Moscow, 2008) [in Russian].
Controlling Flows around Bodies with Vorticity Cells as Applied to Flying Vehicles of Integrated Configuration (Numerical Simulation and Physical Modeling), Ed. by A. V. Ermishin and S. A. Isaev (Mosc. Gos. Univ., Moscow, 2001) [in Russian].
K. N. Volkov, V. N. Emel’yanov, I. V. Teterina, and M. S. Yakovchuk, “Visualization of vortical flows in computational fluid dynamics,” Comput. Math. Math. Phys., 57, 1360–1375 (2017).
ACKNOWLEDGMENTS
This study was supported by the Russian Foundation for Basic Research, project nos. 13-07-12079 and 16-38-60142. The work by A.G. Karpenko was supported by the Russian Foundation for Basic Research, project no. 16-38-60142.
Author information
Authors and Affiliations
Corresponding authors
Additional information
Translated by A. Klimontovich
Rights and permissions
About this article
Cite this article
Volkov, K.N., Emel’yanov, V.N., Karpenko, A.G. et al. Simulating Flows of Viscous Incompressible Fluid on Graphics Processors Using the Splitting Scheme and Multigrid Method. Comput. Math. and Math. Phys. 59, 136–149 (2019). https://doi.org/10.1134/S0965542519010160
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0965542519010160