Parallel FEM Adaptation on Hierarchical Architectures

  • Tomasz Olas
  • Roman Wyrzykowski
  • Pawel Gepner
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7203)


The parallel FEM package NuscaS allows us to solve adaptive FEM problems with 3D unstructured meshes on distributed-memory parallel computers such as PC-clusters. In our previous works, a new method for parallelizing the FEM adaptation was presented, based on using the 8-tetrahedra longest-edge partition. This method relies on a decentralized approach, and is more scalable in comparison to previous implementations requiring a centralized synchronizing node.

At present nodes of clusters contain more and more processing cores. Their efficient utilization is crucial for providing high performance of numerical codes. In this paper, different schemes of mapping the mesh adaptation algorithm on such hierchical architectures are presented and compared. These schemes use either the pure message-passing model, or the hybrid approach which combines shared-memory and message-passing models. Also, we investigate an approach for adapting the pure MPI model to hierarchical topology of clusters with multi-core nodes.


Hierarchical Architecture Mesh Adaptation External Node Neighbor Process Proposed Parallel Algorithm 
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.
    Avera, R., Martino, B., Rak, M., Venticinque, S., Vilano, U.: Performance prediction through simulation of a hybrid MPI/OpenMP application. Parallel Computing 31, 1013–1033 (2005)CrossRefGoogle Scholar
  2. 2.
    Balman, M.: Tetrahedral Mesh Refinement in Distributed Environments. In: 2006 Int. Conf. Parallel Processing Workshops (ICPPW 2006), pp. 497–504. IEEE Computer Soc. (2006)Google Scholar
  3. 3.
    Chorley, M.J., Walker, D.W.: Performance analysis of a hybrod MPI/OpenMP application on multi-core clusters. J. Comput. Sci. 1, 168–174 (2010)CrossRefGoogle Scholar
  4. 4.
    Family of Graph and Hypergraph Partitioning Software,
  5. 5.
    Jeannot, E., Mercier, G.: Improving MPI Applications Performance on Multicore Clusters with Optimized Process Placement. In: 2nd Workshop of COST 0805 Open Network for High-Performance Computing on Complex Environments, Timisoara, January 25-27 (2012)Google Scholar
  6. 6.
    Jin, H., Jespersen, D., Mehrotra, P., Biswas, R., Huang, L., Chapman, B.: High performance computing using MPI and OpenMP on multi-core parallel systems. Parallel Computing 37, 562–575 (2011)CrossRefGoogle Scholar
  7. 7.
    Mercier, G., Clet-Ortega, J.: Towards an Efficient Process Placement Policy for MPI Applications in Multicore Environments. In: Ropo, M., Westerholm, J., Dongarra, J. (eds.) PVM/MPI 2009. LNCS, vol. 5759, pp. 104–115. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  8. 8.
    Olas, T., Karczewski, K., Tomas, A., Wyrzykowski, R.: FEM Computations on Clusters Using Different Models of Parallel Programming. In: Wyrzykowski, R., Dongarra, J., Paprzycki, M., Waśniewski, J. (eds.) PPAM 2001. LNCS, vol. 2328, pp. 170–182. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  9. 9.
    Olas, T., Lacinski, L., Karczewski, K., Tomas, A., Wyrzykowski, R.: Performance of different communication mechanisms for FEM computations on PC-based cluster with SMP nodes. In: Proc. Int. Conf. Parallel Computing in Electrical Engineering, PARELEC 2002, pp. 305–311 (2002)Google Scholar
  10. 10.
    Olas, T., Leśniak, R., Wyrzykowski, R., Gepner, P.: Parallel Adaptive Finite Element Package with Dynamic Load Balancing for 3D Thermo-Mechanical Problems. In: Wyrzykowski, R., Dongarra, J., Karczewski, K., Wasniewski, J. (eds.) PPAM 2009. LNCS, vol. 6067, pp. 299–311. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  11. 11.
    Olas, T., Wyrzykowski, R.: Adaptive FEM Package with Decentralized Parallel Adaptation of Tetrahedral Meshes. In: Lirkov, I. (ed.) LSSC 2011. LNCS, vol. 7116, pp. 622–629. Springer, Heidelberg (in press, 2012)Google Scholar
  12. 12.
    Patzak, B., Rypl, D.: A Framework for Parallel Adaptive Finite Element Computations with Dynamic Load Balancing. In: Proc. First Int. Conf. Parallel, Distributed and Grid Computing for Engineering, Paper 31. Civil-Comp Press (2009)Google Scholar
  13. 13.
    Plaza, A., Rivara, M.: Mesh Refinement Based on the 8-Tetrahedra Longest-Edge Partition. In: Proc. 12th Int. Meshing Roundtable, Sandia National Laboratories, pp. 67–78 (2003)Google Scholar
  14. 14.
    Rivara, M., Pizarro, D., Chrisochoides, N.: Parallel Refinement of Tetrahedral Meshes using Terminal-Edge Bisection Algorithm. In: Proc. 13th Int. Meshing Roundtable, Sandia National Labs, pp. 427–436 (2004)Google Scholar
  15. 15.
    Wei, F., Yilmaz, A.E.: A hybrid message passing/shared memory parallelization of the adaptive integral method for multi-core clusters. Parallel Computing 37, 279–301 (2011)zbMATHCrossRefGoogle Scholar
  16. 16.
    Wyrzykowski, R., Olas, T., Sczygiol, N.: Object-Oriented Approach to Finite Element Modeling on Clusters. In: Sørevik, T., Manne, F., Moe, R., Gebremedhin, A.H. (eds.) PARA 2000. LNCS, vol. 1947, pp. 250–257. Springer, Heidelberg (2001)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Tomasz Olas
    • 1
  • Roman Wyrzykowski
    • 1
  • Pawel Gepner
    • 2
  1. 1.Czestochowa University of TechnologyCzestochowaPoland
  2. 2.Intel CorporationPoland

Personalised recommendations