Advertisement

A New Approach: Component-Based Multi-physics Coupling through CCA-LISI

  • Fang Liu
  • Masha Sosonkina
  • Randall Bramley
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6017)

Abstract

A new problem in scientific computing is the merging of existing simulation models to create new, higher fidelity combined models. This has been a driving force in climate modeling for nearly a decade now, and fusion energy, space weather modeling are starting to integrate different sub-physics into a single model. Through component-based software engineering, an interface supporting this coupling process provides a way to invoke the sub-model through the common interface which the top model uses, then a coupled model turns into a higher level model. In addition to allowing applications to switch among linear solvers, a linear solver interface is also needed for the model coupling. A linear solver interface helps in creating solvers for the integrated multi-physics simulation that combines separate codes, and can use each code’s native and specialized solver for the sub-problem corresponding to each physics sub-model. This paper presents a new approach on coupling multi-physics codes in terms of coupled solver, and shows the successful proof for coupled simulation through the implicit solve.

Keywords

Parallel model coupling Component architecture and interface Sparse matrix computation 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    CCA specification (April 2008), http://www.cca-forum.org/specification
  2. 2.
    Hypre: Scalable Linear Solvers: high performance preconditioners (April 2008), http://www.llnl.gov/CASC/linear_solvers/
  3. 3.
    PETSc: Portable Extensibel Toolkit for Scientific Computation (April 2008), http://www-unix.mcs.anl.gov/petsc/petsc-as/
  4. 4.
    SuperLU: Direct Solver Package of large, sparse, nonsymmetric systems of linear equations (April 2008), http://crd.lbl.gov/~xiaoye/SuperLU/
  5. 5.
    The Trilinos Project (April 2008), http://software.sandia.gov/trilinos
  6. 6.
    Center for simulation of RF wave interactions with magnetohydrodynamics (March 2009), http://cswim.org
  7. 7.
    Community climate system model (March 2009), http://www.ccsm.ucar.edu/
  8. 8.
    Allan, B.A., Armstrong, R.: Ccaffeine Framework: Composing and Debugging Applications Iteratively and Running them Statically. In: Compframe 2005 workshop (June 2005)Google Scholar
  9. 9.
    Allan, B.A., Armstrong, R.C., Wolfe, A.P., Ray, J., Bernholdt, D.E., Kohl, J.A.: The CCA core specification in a distributed memory SPMD framework. Concurrency and Computation: Practice and Experience 14(5), 323–345 (2002)zbMATHCrossRefGoogle Scholar
  10. 10.
    Barrett, R., Berry, M., Chan, T.F., Demmel, J., Donato, J., Dongarra, J., Eijkhout, V., Pozo, R., Romine, C., der Vorst, H.V.: Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods, 2nd edn. SIAM, Philadelphia (1994)Google Scholar
  11. 11.
    Batchelor, D.: Integrated Simulation of Fusion Plasmas. Physics Today 58, 35–40 (2005)CrossRefGoogle Scholar
  12. 12.
    Bertrand, F., Yuan, Y., Chiu, K., Bramley, R.: An approach to parallel MxN communication. In: Proceedings of the Los Alamos Computer Science Institute (LACSI) Symposium, Santa Fe, NM (October 2003)Google Scholar
  13. 13.
    Bjørstad, P.E.: Multiplicative and additive schwarz methods: Convergence in the two-domain case. In: Chan, T., Glowinski, R., Periaux, J., Widlund, O. (eds.) Domain Decomposition Methods, pp. 147–159. SIAM, Philadelphia (1989)Google Scholar
  14. 14.
    CCA-Forum. The DOE Common Component Architecture project (April 2008), http://www.cca-forum.org/
  15. 15.
    Chan, T.F., Resasco, D.C.: A Framework for the Analysis and Construction of Domain Decomposition Preconditioners. In: First International Symposium on Domain Decomposition Methods for Partial Differential EquationsGoogle Scholar
  16. 16.
    Dahlgren, T., Epperly, T., Kumfert, G., Leek, J.: Babel User’s Guide. CASC, Lawrence Livermore National Laboratory, Livermore, CA, babel-0.11.0 edition (2005)Google Scholar
  17. 17.
    Drummond, L.A., Demmel, J., Mechoso, C.R., Robinson, H., Sklower, K., Spahr, J.A.: A data broker for distributed computing environments. In: ICCS 2001: Proceedings of the International Conference on Computational Sciences-Part I, pp. 31–40. Springer, Heidelberg (2001)Google Scholar
  18. 18.
    Dryja, M.: An additive schwarz algorithm for two- and three- dimensional finite element elliptic problems. In: Chan, T., Glowinski, R., Periaux, J., Widlund, O. (eds.) Domain Decomposition Methods, pp. 168–172. SIAM, Philadelphia (1989)Google Scholar
  19. 19.
    Jiao, X., Campbell, M.T., Heath, M.T.: Occom: An object-oriented, data-centric software integration framework for multiphysics simulations. In: ICS, pp. 358–368 (2003)Google Scholar
  20. 20.
    Jill Dahlburg, J.C., et al.: Fusion Simulation Project: Integrated Simulation and Optimization of Magnetic Fusion Systems. Journal of Fusion Energy 20(4), 135–196 (2001), http://www.isofs.info CrossRefGoogle Scholar
  21. 21.
    Kohn, S., Kumfert, G., Painter, J., Ribbens, C.: Divorcing Language Dependencies from a Scientific Software Library. In: 10th SIAM Conference on Parallel Processing. LLNL document UCRL-JC-140349, Portsmouth, VA, March 12-14 (2001), http://www.llnl.gov/CASC/components/babel.html
  22. 22.
    Larson, J., Jacob, R., Ong, E.: The model coupling toolkit: A new fortran90 toolkit for building multiphysics parallel coupled models. International Journal of High Performance Computing Application 19, 277–292 (2005)CrossRefGoogle Scholar
  23. 23.
    Liu, F., Bramley, R.: CCA-LISI: On Designing a CCA Parallel Sparse Linear Solver Interface. In: Proc. 21th Int’l. Parallel & Distributed Processing Symp. (IPDPS), p. 10. ACM/IEEE Computer Society (2007)Google Scholar
  24. 24.
    Liu, F., Sosonkina, M., Bramley, R.: A HPC Sparse Solver Interface for Scalable Multilevel Methods. In: High Performance Computing Symposium, March 22-27 (2009)Google Scholar
  25. 25.
    Saad, Y.: Iterative Methods for Sparse Linear Systems. SIAM, Philadelphia (2003)zbMATHGoogle Scholar
  26. 26.
    Shengsheng, Y., Yuanqiao, W., Liwen, H., Jian, D.: Framework of Distributed Numerical Model Coupling System. In: DS-RT ’05: Proceedings of the 9th IEEE International Symposium on Distributed Simulation and Real-Time Applications, Washington, DC, USA, pp. 187–194. IEEE Computer Society Press, Los Alamitos (2005)CrossRefGoogle Scholar
  27. 27.
    Sosonkina, M., Liu, F., Bramley, R.: Usability Levels for Sparse Linear Algebra Components. Concurrency and Computation: Practice and Experience 20, 1439–1454 (2008)CrossRefGoogle Scholar
  28. 28.
    Szyperski, C.: Component Software: Beyong Object-Oriented Programming. ACM Press, New York (1999)Google Scholar
  29. 29.
    Toth, G., Volberg, O., et al.: A Physics-Based Software Framework for Sun-Earth Connection Modeling. In: Liu, A.T.Y., Kamide, Y., Consolini, G. (eds.) Multiscale Coupling of Sun-Earth Processes, Proceeding of the Conference on the Sun-Earth Connection, pp. 383–397 (2004)Google Scholar
  30. 30.
    Zhang, F.: The Schur Complement and Its Application. In: Numerical Methods and Algorithms., Springer, Heidelberg (2005)Google Scholar
  31. 31.
    Zhou, S.J.: Coupling climate models with the earth system modeling framework and the common component architecture. Concurr. Comput.: Pract. Exper. 18(2), 203–213 (2006)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Fang Liu
    • 1
  • Masha Sosonkina
    • 1
  • Randall Bramley
    • 2
  1. 1.Scalable Computing LabUSDOE Ames LaboratoryAmesU.S.A
  2. 2.Computer Science DepartmentIndiana University-BloomingtonBloomingtonU.S.A

Personalised recommendations