Advertisement

The HTFETI Method Variant Gluing Cluster Subdomains by Kernel Matrices Representing the Rigid Body Motions

  • Alexandros MarkopoulosEmail author
  • Lubomír ŘíhaEmail author
  • Tomáš Brzobohatý
  • Ondřej Meca
  • Radek Kučera
  • Tomáš Kozubek
Conference paper
Part of the Lecture Notes in Computational Science and Engineering book series (LNCSE, volume 125)

Abstract

The proposed algorithm called the Hybrid Total Finite Element Tearing and Interconnecting method (HTFETI) is a variant of the TFETI domain decomposition method suitable for large-scale problems with hundreds of thousands of subdomains. The floating subdomains are gathered into several groups belonging to individual clusters. We use the new idea consisting in gluing the cluster subdomains using kernel matrices defined by the rigid body motions. This technique reduces the size of the coarse problem. While the size of the coarse problem depends linearly on the number of subdomains in the classical TFETI method, it depends linearly on the number of clusters in the HTFETI method. The zero weighted averages across the interfaces of neighbouring subdomains (an alternative to the constraints enforcing the continuity across the corners used, e.g., in the FETI-DP method) improve conditioning of the resulting system of linear equations.

Notes

Acknowledgements

This work was supported by The Ministry of Education, Youth and Sports from the National Programme of Sustainability (NPU II) project “IT4Innovations excellence in science - LQ1602” and from the Large Infrastructures for Research, Experimental Development and Innovations project “IT4Innovations National Supercomputing Center – LM2015070”.

References

  1. 1.
    T. Brzobohatý, M. Jarošová, T. Kozubek, M. Menšík, A. Markopoulos, The hybrid total FETI method, in Proceedings of the Third International Conference on Parallel, Distributed, Grid and Cloud Computing for Engineering (Civil-Comp. Ltd, 2011)Google Scholar
  2. 2.
    Z. Dostál, D. Horák, R. Kučera, Total FETI - an easier implementable variant of the FETI method for numerical solution of elliptic PDE. Commun. Numer. Methods Eng. 196, 1155–1162 (2006)MathSciNetCrossRefGoogle Scholar
  3. 3.
    ESPRESO - Exascale Parallel FETI Solver, http://espreso.it4i.cz
  4. 4.
    C. Farhat, F.X. Roux, J.T. Oden, Implicit Parallel Processing in Structural Mechanics (Elsevier, New York, 1994)zbMATHGoogle Scholar
  5. 5.
    C. Farhat, M. Lesoinne, P. LeTallec, K. Pierson, D. Rixen, FETI-DP: a dual-primal unified FETI method, part I: a faster alternative to the two-level FETI method. Int. J. Numer. Methods Eng. 50(7), 1523–1544 (2001)CrossRefGoogle Scholar
  6. 6.
    A. Klawonn, R. Rheinbach, Highly scalable parallel domain decomposition methods with an application to biomechanics. Z. Angew. Math. Mech. 1, 5–32 (2010)MathSciNetCrossRefGoogle Scholar
  7. 7.
    A. Klawonn, O.B. Widlund, M. Dryja, Dual-primal FETI methods for three-dimensional elliptic problems with heterogeneous coefficients. SIAM J. Numer. Anal. 40, 159–179 (2002)MathSciNetCrossRefGoogle Scholar
  8. 8.
    A. Klawonn, M. Lanser, O. Rheinbach, Toward extremely scalable nonlinear domain decomposition methods for elliptic partial differential equations. SIAM J. Sci. Comput. 37(6), C667–C696 (2015). https://doi.org/10.1137/140997907 MathSciNetCrossRefGoogle Scholar
  9. 9.
    R. Kučera, T. Kozubek, A. Markopoulos, On large-scale generalized inverses in solving two-by-two block linear systems. Linear Algebra Appl. 438(7), 3011–3029 (2013)MathSciNetCrossRefGoogle Scholar
  10. 10.
    M. Merta, L. Riha, O. Meca, A. Markopoulos, T. Brzobohaty, T. Kozubek, V. Vondrak, Intel Xeon Phi acceleration of hybrid total FETI solver. Adv. Eng. Softw. n112(Suppl. C), 124–135 (2017). http://www.sciencedirect.com/science/article/pii/S0965997816302745
  11. 11.
    L. Říha, T. Brzobohatý, A. Markopoulos, O. Meca, T. Kozubek, Massively parallel hybrid total FETI (HTFETI) solver, in Proceedings of the Platform for Advanced Scientific Computing Conference. PASC ’16 (ACM, New York, 2016), pp. 7:1–7:11. http://doi.acm.org/10.1145/2929908.2929909

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Alexandros Markopoulos
    • 1
    Email author
  • Lubomír Říha
    • 1
    Email author
  • Tomáš Brzobohatý
    • 1
  • Ondřej Meca
    • 1
  • Radek Kučera
    • 1
  • Tomáš Kozubek
    • 1
  1. 1.IT4Innovations National Supercomputing CenterVŠB - Technical University of OstravaOstravaCzech Republic

Personalised recommendations