DOI: https://doi.org/10.26089/NumMet.v17r331

Efficiency analysis of hydrodynamic calculations on GPU and CPU clusters

Authors

  • A.V. Sentyabov
  • A.A. Gavrilov
  • M.A. Krivov
  • A.A. Dekterev
  • M.N. Pritula

Keywords:

GPGPU
numerical simulation
computational fluid dynamics
SIMPLE
MPI
CUDA

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.

New computational technologies

Downloads

Published

2016-08-12

Issue

Vol. 17 (2016): Issue 3.

Section

Section 1. Numerical methods and applications

Author Biographies

A.V. Sentyabov

Kutateladze Institute of Thermophysics of SB RAS
• Junior Researcher

A.A. Gavrilov

Kutateladze Institute of Thermophysics of SB RAS
• Lead Engineer

M.A. Krivov

ttgLabs, LLC

A.A. Dekterev

Kutateladze Institute of Thermophysics of SB RAS
• Senior Researcher

M.N. Pritula

ttgLabs, LLC
• Developer

References

  1. Supercomputing News, Parallel.ru.
    http://parallel.ru/news/top500_47edition.html . Cited August 30, 2016.
  2. Top 500.
    https://www.top500.org/lists/2016/06/. Cited August 30, 2016.
  3. Supercomputer Center of Moscow University.
    http://parallel.ru/cluster . Cited August 30, 2016.
  4. D. A. Bikulov and D. S. Senin, “Implementation of the Lattice Boltzmann Method without Stored Distribution Functions on GPU,” Vychisl. Metody Programm. 14, 370-374 (2013).
  5. A. N. Maslii and E. I. Madirov, “Performance Comparison of Quantum-Chemical Calculations Using GPU,” Vestn. Kazan Tekhnol. Univ. 16 (23), 12-18 (2013).
  6. A. A. Yunusov, I. M. Gubaydullin, and M. R. Fayzullin, “Analysis of Algorithms for Solving Chemical Kinetics Problems Using GPGPU,” Zh. Srednevolzh. Matem. Obshchestva 12 (3), 146-152 (2010).
  7. A. V. Gorobets, S. A. Sukov, A. O. Zheleznyakov, et al., “Application of GPU for Hybrid Two-Level Parallelization MPI+OpenMP on Heterogeneous Computing Systems,” in Proc. Int. Conf. on Parallel Computational Technologies, Moscow, Russia, March 28-April 1, 2011 (South Ural State Univ., Chelyabinsk, 2011), pp. 452-460.
  8. K. N. Volkov, Yu. N. Deryugin, V. N. Emel’yanov, A. G. Karpenko, A. S. Kozelkov, and I. V. Teterina, Methods for Accelerating Gasdynamic Calculations on Unstructured Grids (Fizmatlit, Moscow, 2013) [in Russian].
  9. A. A. Gavrilov, A. V. Minakov, A. A. Dekterev, and V. Ya. Rudyak, “A Numerical Algorithm for Modeling Laminar Flows in an Annular Channel with Eccentricity,” Sib. Zh. Ind. Mat. 13 (4), 3-14 (2010) [J. Appl. Ind. Math. 5 (4), 559-568 (2011)].
  10. A. A. Gavrilov, A. V. Minakov, A. A. Dekterev, and V. Ya. Rudyak, “Numerical Algorithm for Fully Developed Laminar Flow of a Non-Newtonian Fluid through an Eccentric Annulus,” Vychisl. Tekhnol. 17 (1), 44-56 (2012).
  11. D. J. Mavriplis, “Revisiting the Least-Squares Procedure for Gradient Reconstruction on Unstructured Meshes,” AIAA Paper 2003-3986 (2003).
  12. J. H. Ferziger and M. Peric, Computational Methods for Fluid Dynamics (Springer, Heidelberg, 2002).
  13. F. Moukalled and M. Darwish, “A Unified Formulation of the Segregated Class of Algorithms for Fluid Flow at All Speeds,” Numer. Heat Transfer. Part B. 37 (2), 227-246 (2000).
  14. I. A. Belov, S. A. Isaev, and V. A. Korobkov, Problems and Methods of Calculation of Separating Flows of Incompressible Fluids (Sudostroenie, Leningrad, 1989) [in Russian].
  15. S. Patankar, Numerical Heat Transfer and Fluid Flow (Hemisphere, New York, 1980; Energoatomizdat, 1984).
  16. C. M. Rhie and W. L. Chow, “Numerical Study of the Turbulent Flow Past an Airfoil with Trailing Edge Separation,” AIAA J. 21 (11), 1525-1532 (1983).
  17. B. P. Leonard, “A Stable and Accurate Convective Modelling Procedure Based on Quadratic Upstream Interpolation,” Comput. Methods Appl. Mech. Eng. 19 (1), 59-98 (1979).
  18. R. Barrett, M. W. Berry, T. F. Chan, et al., Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods (SIAM, Philadelphia, 1994).
  19. A. A. Samarskii and E. S. Nikolaev, Numerical Methods for Grid Equations (Nauka, Moscow, 1978; Birkh854user, Basel, 1989).
  20. G. Karypis and V. Kumar, “A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs,” SIAM J. Sci. Comput. 20 (1), 359-392 (1998).
  21. A. A. Gavrilov, M. A. Krivov, S. A. Grizan, and A. A. Dekterev, “GPU Version of CFD Software SigmaFlow: Porting and Optimization Using TTG Apptimizer Toolkit,” in Proc. Int. Conf. on Parallel Computational Technologies, Chelyabinsk, Russia, April 1-5, 2013 (South Ural State Univ., Chelyabinsk, 2013), pp. 106-115.
  22. Vl. V. Voevodin, S. A. Zhumatii, S. I. Sobolev, et al., “The Lomonosov Supercomputer in Practice,” Otkrytye Sistemy, No. 7, 36-39 (2012).
  23. V. Sadovnichy, A. Tikhonravov, Vl. Voevodin, and V. Opanasenko, “’Lomonosov’: Supercomputing at Moscow State University,” in Contemporary High Performance Computing: From Petascale toward Exascale (CRC Press, Boca Raton, 2013), pp. 283-307.