Experimental Optimization of Parallel 3D Overlapping Domain Decomposition Schemes

  • Sofia Guzzetti
  • Alessandro Veneziani
  • Vaidy Sunderam
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9573)


Overlapping domain decomposition is a possible convenient technique for solving complex problems described by Partial Differential Equations in a parallel framework. The performance of this approach strongly depends on the size and the position of the overlap since the overlapping has a positive impact on the number of iterations required by the numerical scheme and the relatively flexible and judicious choice of the interface may lead to a reduction of the communication time. In this paper we test the overlapping domain decomposition method on the finite element discretization of a diffusion reaction problem in both idealized and real 3D geometries. Results confirm that the detection of the optimal overlapping in real cases is not trivial but has the potential to significantly reduce the computational costs of the entire solution process.


Applied computing in medicine Computational fluid dynamics Parallel overlapping domain decomposition Performance analysis 


  1. 1.
    Antiga, L., Passerini, T., Piccinelli, M., Veneziani, A.: Aneurisk web. http://ecm2.mathcs.emory.edu/aneuriskweb/index
  2. 2.
    Cai, X.: Overlapping domain decomposition methods. In: Langtangen, H.P., Tveito, A. (eds.) Advanced Topics in Computational Partial Differential Equations, pp. 57–95. Springer, Heidelberg (2003)Google Scholar
  3. 3.
    Catalyurek, U.V., Aykanat, C.: Hypergraph-partitioning-based decomposition for parallel sparse-matrix vector multiplication. IEEE Trans. Parallel Distrib. Syst. 10(7), 673–693 (1999)CrossRefGoogle Scholar
  4. 4.
    Darjany, D., Englert, B., Kim, E.H.: Implementing overlapping domain decomposition methods on a virtual parallel machine. In: Min, G., Di Martino, B., Yang, L.T., Guo, M., Rünger, G. (eds.) ISPA Workshops 2006. LNCS, vol. 4331, pp. 717–727. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  5. 5.
    Formaggia, L., Saleri, F., Veneziani, A.: Solving Numerical PDEs: Problems, Applications, Exercises: Problems, Applications, Exercises. Springer Science & Business Media, New York (2012)CrossRefMATHGoogle Scholar
  6. 6.
  7. 7.
    Quarteroni, A., Valli, A.: Domain decomposition methods for partial differential equations. Technical report. Oxford University Press (1999)Google Scholar
  8. 8.
    Rahman, T., Valdman, J.: Fast MATLAB assembly of FEM matrices in 2D and 3D: nodal elements. Appl. Math. Comput. 219(13), 7151–7158 (2013)MathSciNetMATHGoogle Scholar
  9. 9.
    Saad, Y.: Iterative Methods for Sparse Linear Systems. Siam, Philadelphia (2003)CrossRefMATHGoogle Scholar
  10. 10.
    Selvitopi, R.O., Turk, A., Aykanat, C.: Replicated partitioning for undirected hypergraphs. J. Parallel Distrib. Comput. 72(4), 547–563 (2012)CrossRefGoogle Scholar
  11. 11.
    Slawinski, J.: Adaptive Approaches to Utility Computing for Scientific Applications. Ph.D. thesis, Emory University (2014)Google Scholar
  12. 12.
    Slawinski, J., Passerini, T., Villa, U., Veneziani, A., Sunderam, V.: Experiences with target-platform heterogeneity in clouds, grids, and on-premises resources. In: 2012 IEEE 26th International Parallel and Distributed Processing Symposium Workshops & PhD Forum (IPDPSW), pp. 41–52 (2012)Google Scholar
  13. 13.
    Toselli, A., Widlund, O.: Domain Decomposition Methods: Algorithms and Theory. Springer Series in Computational Mathematics, vol. 34. Springer, Heidelberg (2005)CrossRefMATHGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Sofia Guzzetti
    • 1
  • Alessandro Veneziani
    • 1
  • Vaidy Sunderam
    • 1
  1. 1.Department of Mathematics and Computer ScienceEmory UniversityAtlantaUSA

Personalised recommendations