DOI: https://doi.org/10.26089/NumMet.v19r211

Acceleration of parallel algorithms for solving three-dimensional boundary value problems on quasi-structured grids

Authors

  • I.A. Klimonov
  • V.D. Korneev
  • V.M. Sveshnikov

Keywords:

boundary value problems
parallelization
quasi-structured grids
iterative process
initial approximation

Abstract

This paper is devoted to the acceleration of the parallel solution of three-dimensional boundary value problems by the computational domain decomposition method into subdomains that are conjugated without overlapping. The decomposition is performed by a uniform parallelepipedal macrogrid. In each subdomain and on the interface, some structured subgrids are constructed. The union of these subgrids forms a quasi-structured grid on which the problem is solved. The parallelization is carried out using the MPI-technology. We propose and experimentally study the acceleration algorithm for an external iterative process on subdomains to solve a system of linear algebraic equations approximating the Poincare-Steklov equation on the interface. A number of numerical experiments are carried out on various quasi-structured grids and with various parameters of computational algorithms showing the acceleration of computations.

New computational technologies

Downloads

Published

2018-03-30

Issue

Vol. 19 (2018): Issue 2.

Section

Section 1. Numerical methods and applications

Author Biographies

I.A. Klimonov

Novosibirsk State University
• PhD Student

V.D. Korneev

The Institute of Computational Mathematics and Mathematical Geophysics of SB RAS (ICM&MG SB RAS)
• Senior Researcher

V.M. Sveshnikov

The Institute of Computational Mathematics and Mathematical Geophysics of SB RAS (ICM&MG SB RAS)
• Head of Laboratory

References

  1. V. D. Korneev and V. M. Sveshnikov, “Parallel Algorithms and Domain Decomposition Techniques for Solving Three-Dimensional Boundary Value Problems on Quasi-Structured Grids,” Sib. Zh. Vych. Mat. 19 (2), 183-194 (2016) [Numer. Anal. Appl. 9 (2), 141-149 (2016)].
  2. I. A. Klimonov, V. D. Korneev, and V. M. Sveshnikov, “Parallelization Technologies for Solving Three-Dimensional Boundary Value Problems on Quasi-Structured Grids Using the CPU+GPU Hybrid Computing Environment,” Vychisl. Metody Programm. 17, 65-71 (2016).
  3. A. Quarteroni and A. Valli, Domain Decomposition Methods for Partial Differential Equations (Clarendon Press, Oxford, 1999).
  4. V. Dolean, P. Jolivet, and F. Nataf, An Introduction to Domain Decomposition Methods: Algorithms, Theory, and Parallel Implementation (SIAM Press, Philadelphia, 2015).
  5. A. A. Samarskii, The Theory of Difference Schemes (Nauka, Moscow, 1977; Marcel Dekker, New York, 2001).
  6. V. P. Il’in, Finite Difference and Finite Volume Methods for Elliptic Equations (Inst. Comput. Math. Math. Geophys., Novosibirsk, 2000) [in Russian].
  7. V. P. Il’in, Methods and Technologies of Finite Elements (Inst. Comput. Math. Math. Geophys., Novosibirsk, 2007) [in Russian].
  8. A. A. Samarskii and A. V. Gulin, Numerical Methods of Mathematical Physics (Nauchnyi Mir, Moscow, 2003) [in Russian].