DOI: https://doi.org/10.26089/NumMet.v16r110

Domain decomposition based on a direct method for solving the three-dimensional Poisson’s equation in nonstationary astrophysical problems

Authors

  • N.V. Snytnikov

Keywords:

Poisson’s equation
Dirichlet problem
domain decomposition
gravitational potential
stellar dynamics
parallel programming
scalability of algorithms

Abstract

A new parallel algorithm for solving the three-dimensional Poisson’s equation in the context of nonstationary problems of astrophysics is proposed. This algorithm is based on a decomposition of the 3D domain in two directions, on the application of a direct method for solving the Dirichlet problem in each subdomain, and on a combination of subdomains coupling for the screened Poisson’s equation with the variable separation method. Test experiments were conducted on supercomputers installed at the Joint Supercomputing Center of Russian Academy of Sciences (Moscow) and at the Siberian Supercomputing Center (Novosibirsk).

New computational technologies

Downloads

Published

2015-03-03

Issue

Vol. 16 (2015): Issue 1.

Section

Section 1. Numerical methods and applications

Author Biography

N.V. Snytnikov

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

References

  1. V. A. Vshivkov, V. N. Snytnikov, and N. V. Snytnikov, “Simulation of the Three-Dimensional Dynamics of Matter in the Gravitational Field Using Multiprocessor Computers,” Vychisl. Tekhnol. 11 (2), 15-27 (2006).
  2. V. A. Vshivkov, G. G. Lazareva, A. V. Snytnikov, et al., “Hydrodynamical Code for Numerical Simulation of the Gas Components of Colliding Galaxies,” Astrophys. J. Suppl. Ser. 194 (2), 1-12 (2011).
  3. V. Springel, N. Yoshida, and S. D. M. White, “GADGET: A Code for Collisionless and Gasdynamical Cosmological Simulation,” New Astr. 6 (2), 79-117 (2001).
  4. V. N. Snytnikov, V. A. Vshivkov, E. A. Kuksheva, et al., “Three-Dimensional Numerical Simulation of a Nonstationary Gravitating N-Body System with Gas,” Pis’ma v Astron. Zh. 30 (2), 146-160 (2004) [Astron. Lett. 30 (2), 124-137 (2004)].
  5. V. A. Vshivkov and A. V. Snytnikov, “Development of an Efficient Parallel Poisson Equation Solver for the Simulation of Protoplanetary Disk Evolution,” Vychisl. Metody Programm. 10, 116-122 (2009).
  6. N. V. Snytnikov, “A Parallel Algorithm for Solving 2D Poisson’s Equation in the Context of Nonstationary Problems,” Vychisl. Metody Programm. 16, 39-51 (2015).
  7. J. Huang and L. Greengard, “A Fast Direct Solver for Elliptic Partial Differential Equations on Adaptively Refined Meshes,” SIAM J. Sci. Comput. 21 (4), 1551-1566 (1999).
  8. A. V. Terekhov, “Parallel Dichotomy Algorithm for Solving Tridiagonal System of Linear Equations with Multiple Right-Hand Sides,” Parallel Comput. 36 (8), 423-438 (2010).
  9. O. Ayala and L.-P. Wang, “Parallel Implementation and Scalability Analysis of 3D Fast Fourier Transform Using 2D Domain Decomposition,” Parallel Comput. 39 (1), 58-77 (2013).
  10. T. V. T. Duy and T. Ozaki, “A Decomposition Method with Minimum Communication Amount for Parallelization of Multi-Dimensional FFTs,” Comput. Phys. Commun. 185 (1), 153-164 (2014).
  11. Intel Math Kernel Library 10.0: Overview.
    https://software.intel.com/en-us/intel-mkl . Cited January 28, 2015.
  12. M. Frigo and S. G. Johnson, “FFTW software,”
    http://www.fftw.org . Cited January 28, 2015.