Advertisement

A High-Performance Implementation of a Robust Preconditioner for Heterogeneous Problems

  • Linus SeelingerEmail author
  • Anne Reinarz
  • Robert Scheichl
Conference paper
  • 42 Downloads
Part of the Lecture Notes in Computer Science book series (LNCS, volume 12043)

Abstract

We present an efficient implementation of the highly robust and scalable GenEO (Generalized Eigenproblems in the Overlap) preconditioner [16] in the high-performance PDE framework DUNE [6]. The GenEO coarse space is constructed by combining low energy solutions of a local generalised eigenproblem using a partition of unity. The main contribution of this paper is documenting the technical details that are crucial to the efficiency of a high-performance implementation of the GenEO preconditioner. We demonstrate both weak and strong scaling for the GenEO solver on over 15, 000 cores by solving an industrially motivated problem in aerospace engineering. Further, we show that for highly complex parameter distributions arising in certain real-world applications, established methods become intractable while GenEO remains fully effective.

Keywords

Partial differential equations Domain decomposition Preconditioning High performance computing 

Notes

Acknowledgements

This work was supported by an EPSRC Maths for Manufacturing grant (EP/K031368/1). This research made use of the Balena High Performance Computing Service at the University of Bath. This work used the ARCHER UK National Supercomputing Service (http://www.archer.ac.uk).

References

  1. 1.
    Alnæs, M.S., et al.: The FEniCS project version 1.5. Arch. Numer. Softw. 3(100), 9–23 (2015).  https://doi.org/10.11588/ans.2015.100.20553CrossRefGoogle Scholar
  2. 2.
    Alzetta, G., et al.: The deal.II library version 9.0. J. Numer. Math. 26(4), 173–183 (2018).  https://doi.org/10.1515/jnma-2018-0054MathSciNetCrossRefzbMATHGoogle Scholar
  3. 3.
    Babuška, I., Caloz, G., Osborn, J.E.: Special finite element methods for a class of second order elliptic problems with rough coefficients. SIAM J. Numer. Anal. 31(4), 945–981 (1994)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Bastian, P., Blatt, M.: On the generic parallelisation of iterative solvers for the finite element method. Int. J. Comput. Sci. Eng. 4(1), 56–69 (2008)Google Scholar
  5. 5.
    Bastian, P., et al.: A generic grid interface for parallel and adaptive scientific computing. Part ii. Implementation and tests in dune. Computing 82(2–3), 121–138 (2008)MathSciNetCrossRefGoogle Scholar
  6. 6.
    Bastian, P., Heimann, F., Marnach, S.: Generic implementation of finite element methods in the distributed and unified numerics environment (DUNE). Kybernetika 46(2), 294–315 (2010)MathSciNetzbMATHGoogle Scholar
  7. 7.
    Butler, R., Dodwell, T., Reinarz, A., Sandhu, A., Scheichl, R., Seelinger, L.: Dune-composites - an open source, high performance package for solving large-scale anisotropic elasticity problems. arXiv e-prints arXiv:1901.05188 (January 2019)
  8. 8.
    Chung, E., Efendiev, Y., Tat Leung, W., Ye, S.: Generalized multiscale finite element methods for space-time heterogeneous parabolic equations. Comput. Math. Appl. 76(2), 419–437 (2016).  https://doi.org/10.1016/j.camwa.2018.04.028MathSciNetCrossRefzbMATHGoogle Scholar
  9. 9.
    Jolivet, P., Hecht, F., Nataf, F., Prud’homme, C.: Scalable domain decomposition preconditioners for heterogeneous elliptic problems. In: Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, pp. 80:1–80:11. SC 2013. ACM, New York (2013).  https://doi.org/10.1145/2503210.2503212
  10. 10.
    Lehoucq, R.B., Sorensen, D.C., Yang, C.: ARPACK users guide: solution of large scale eigenvalue problems by implicitly restarted Arnoldi methods (1997)Google Scholar
  11. 11.
    Pechstein, C., Dohrmann, C.R.: A unified framework for adaptive BDDC. Electron. Trans. Numer. Anal. 46, 273–336 (2017)MathSciNetzbMATHGoogle Scholar
  12. 12.
    Reinarz, A., Dodwell, T., Fletcher, T., Seelinger, L., Butler, R., Scheichl, R.: Dune-composites - a new framework for high-performance finite element modelling of laminates. Compos. Struct. 184, 269–278 (2018)CrossRefGoogle Scholar
  13. 13.
    Sandhu, A., Reinarz, A., Dodwell, T.: A bayesian framework for assessing the strength distribution of composite structures with random defects. Compos. Struct. 205, 58–68 (2018).  https://doi.org/10.1016/j.compstruct.2018.08.074CrossRefGoogle Scholar
  14. 14.
    Smith, B.F., Bjørstad, P.E., Gropp, W.: Domain Decomposition. Cambridge University Press, Cambridge (1996). includes bibliographical referenceszbMATHGoogle Scholar
  15. 15.
    Spillane, N., Dolean, V., Hauret, P., Nataf, F., Pechstein, C., Scheichl, R.: Abstract robust coarse spaces for systems of PDEs via generalized eigenproblems in the overlaps. Numer. Math. 126(4), 741–770 (2014).  https://doi.org/10.1007/s00211-013-0576-yMathSciNetCrossRefzbMATHGoogle Scholar
  16. 16.
    Spillane, N., Dolean, V., Hauret, P., Nataf, F., Pechstein, C., Scheichl, R.: A robust two-level domain decomposition preconditioner for systems of PDEs. C. R. Math. 349(23–24), 1255–1259 (2011)MathSciNetCrossRefGoogle Scholar
  17. 17.
    Toselli, A., Widlund, O.: Domain Decomposition Methods - Algorithms and Theory. Springer Series in Computational Mathematics. Springer, Heidelberg (2005).  https://doi.org/10.1007/b137868CrossRefzbMATHGoogle Scholar
  18. 18.
    Yang, U.M., Henson, V.E.: BoomerAMG: a parallel algebraic multigrid solver and preconditioner. Appl. Numer. Math. 41(1), 155–177 (2002)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Institute for Scientific ComputingHeidelberg UniversityHeidelbergGermany
  2. 2.Department of InformaticsTechnical University of MunichGarchingGermany
  3. 3.Institute for Applied MathematicsHeidelberg UniversityHeidelbergGermany

Personalised recommendations