Advertisement

Highly Parallel Multigrid Solvers for Multicore and Manycore Processors

  • Oleg Bessonov
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9251)

Abstract

In this paper we present an analysis of parallelization properties and implementation details of the new Algebraic multigrid solvers. Variants of smoothers and multicolor grid partitionings are discussed. Optimizations for modern throughput-oriented processors are considered together with different storage schemes. Finally, comparative performance results for multicore and manycore processors are presented.

Notes

Acknowledgements

This work was supported by the Russian Foundation for Basic Research (project RFBR-15-01-06363) and by the Institute of mathematics (IMATH) of the University of Toulon. Computations have been performed at the BULL’s Computing Center, IMATH and Mésocentre of the University of Aix-Marseille, France.

References

  1. 1.
    Shewchuk, J.R.: An Introduction to the Conjugate Gradient Method without the Agonizing Pain. Carnegie Mellon University, School of Computer Science, Pittsburgh (1994)Google Scholar
  2. 2.
    Accary, G., Bessonov, O., Fougère, D., Gavrilov, K., Meradji, S., Morvan, D.: Efficient parallelization of the preconditioned conjugate gradient method. In: Malyshkin, V. (ed.) PaCT 2009. LNCS, vol. 5698, pp. 60–72. Springer, Heidelberg (2009) CrossRefGoogle Scholar
  3. 3.
    Bessonov, O.: Parallelization properties of preconditioners for the conjugate gradient methods. In: Malyshkin, V. (ed.) PaCT 2013. LNCS, vol. 7979, pp. 26–36. Springer, Heidelberg (2013) CrossRefGoogle Scholar
  4. 4.
    Stüben, K.: A review of algebraic multigrid. J. Comput. Appl. Math. 128, 281–309 (2001)MathSciNetCrossRefzbMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  1. 1.Institute for Problems in Mechanics of the Russian Academy of SciencesMoscowRussia

Personalised recommendations