Balancing the Communications and Computations in Parallel FEM Simulations on Unstructured Grids

  • Nikola Kosturski
  • Svetozar Margenov
  • Yavor Vutov
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7204)


We consider large scale finite element modeling on 3D unstructured grids. Large scale problems imply the use of parallel hardware and software. In general, the computational process on unstructured grids includes: mesh generation, mesh partitioning, optional mesh refinement, discretization, and the solution. The impact of the domain partitioning strategy on the performance of the discretization and solution stages is studied.

Our investigations are focused on the Blue Gene/P massively parallel computer. The mapping of the communications to the underlying 3D tours interconnect topology is considered as well.

As a sample problem, we consider the simulation of the thermal and electrical processes, involved in the radio-frequency (RF) ablation procedure. RF ablation is a low invasive technique for the treatment of hepatic tumors, utilizing AC current to destroy the tumor cells by heating.


Unstructured Grid Linear Solver Torus Network Large Scale Finite Element Sparse Matrix Block 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Karypis, G., Kumar, V.: Multilevel k-way partitioning scheme for irregular graphs. Journal of Parallel and DistributedComputing 48(1) (1998)Google Scholar
  2. 2.
    Karypis, G., Kumar, V.: A coarse-grain parallel multilevel k-way partitioning algorithm. In: Proceedings of the 8th SIAM Conference on Parallel Processing for Scientific Computing (1997)Google Scholar
  3. 3.
    ParMETIS - Parallel Graph Partitioning and Fill-reducing Matrix Ordering,
  4. 4.
    Henson, V.E., Yang, U.M.: BoomerAMG: A parallel algebraic multigrid solver and preconditioner. Applied Numerical Mathematics 41(1), 155–177 (2002)MathSciNetzbMATHCrossRefGoogle Scholar
  5. 5.
    Pellegrini, F., Roman, J.: Experimental analysis of the dual recursive bipartitioning algorithm for static mapping. Research Report, LaBRI, Universit e Bordeaux I (August 1996),
  6. 6.
    Scotch and PT-Scotch: Software package and libraries for sequential and parallel graph partitioning, static mapping, and sparse matrix block ordering, and sequential mesh and hypergraph partitioning,
  7. 7.
    Tungjitkusolmun, S., Staelin, S.T., Haemmerich, D., Tsai, J.Z., Cao, H., Webster, J.G., Lee, F.T., Mahvi, D.M., Vorperian, V.R.: Three-dimensional finite-element analyses for radio-frequency hepatic tumor ablation. IEEE Transactions on Biomedical Engineering 49(1), 3–9 (2002)CrossRefGoogle Scholar
  8. 8.
    Tungjitkusolmun, S., Woo, E.J., Cao, H., Tsai, J.Z., Vorperian, V.R., Webster, J.G.: Thermal-electrical finite element modelling for radio frequency cardiac ablation: Effects of changes in myocardial properties. Medical and Biological Engineering and Computing 38(5), 562–568 (2000)CrossRefGoogle Scholar
  9. 9.
    Axelsson, O.: Iterative Solution Methods. Cambridge University Press (1996)Google Scholar
  10. 10.
    Brenner, S., Scott, L.: The mathematical theory of finite element methods. Texts in Applied Mathematics, vol. 15. Springer, Heidelberg (1994)zbMATHGoogle Scholar
  11. 11.
    Hairer, E., Norsett, S.P., Wanner, G.: Solving ordinary differential equations I, II. Springer Series in Comp. Math. (2002)Google Scholar
  12. 12.
    Kosturski, N., Margenov, S.: Supercomputer Simulation of Radio-Frequency Hepatic Tumor Ablation, AMiTaNS 2010 Proceedings. AIP CP, vol. 1301, pp. 486–493 (2010)Google Scholar
  13. 13.
    Lawrence Livermore National Laboratory, Scalable Linear Solvers Project,
  14. 14.

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Nikola Kosturski
  • Svetozar Margenov
  • Yavor Vutov

There are no affiliations available

Personalised recommendations