Skip to main content

Fragmentation of numerical algorithms for parallel subroutines library

Abstract

Fragmentation is a well-known method of the parallelization of numerical algorithms and programs. Algorithm fragmentation allows creating fragmented parallel programs that can be executed on parallel computers of different types (multiprocessors and/or multicomputers) and can be dynamically tuned to all the available resources. Fragmentation of the often used numerical algorithms, their representation for inclusion into the library of parallel numerical subroutines and properties of the runtime system are considered.

This is a preview of subscription content, access via your institution.

References

  1. 1.

    Glushkov VM, Ignatyev MV, Myasnikov VA, Torgashev VA (1974) Recursive machines and computing technologies. In: Proceedings of the IFIP congress, vol 1. North-Holland, Amsterdam, pp 65–70

    Google Scholar 

  2. 2.

    Hill J, McColl W, Stefanescu D, Goudreau M, Lang K, Rao S, Suel T, Tsantilas Th, Bisseling R (1998) BSPlib: The BSP programming library. Parallel Comput 24:1947–1980

    Article  Google Scholar 

  3. 3.

    Torgashev VA, Tsarev IV (2001) Sredstva organozatsii parallelnykh vychislenii i programmirovaniya v multiprocessorakh s dynamicheskoi architechturoi. Programmirovanie 4:53–67

    Google Scholar 

  4. 4.

    BSPlib, http://www.bsp-worldwide.org/

  5. 5.

    BLAS, http://www.netlib.org/blas/

  6. 6.

    ScaLAPACK, http://www.netlib.org/scalapack/

  7. 7.

    Intel MKL, http://software.intel.com/en-us/articles/intel-math-kernel-library-documentation/

  8. 8.

    Charm++, http://charm.cs.uiuc.edu/

  9. 9.

    Cell Superscalar, http://www.bsc.es/cellsuperscalar

  10. 10.

    SMP Superscalar, http://www.bsc.es/smpsuperscalar

  11. 11.

    Grid Superscalar, http://www.bsc.es/grid/grid_superscalar

  12. 12.

    ALF, http://www.ibm.com/developerworks/power/cell/documents.html.

  13. 13.

    Kraeva MA, Malyshkin VE (2001) Assembly technology for parallel realization of numerical models on MIMD-multicomputers. Int J Future Gener Comput Syst 17(6):755–765

    MATH  Article  Google Scholar 

  14. 14.

    Valkovskii V, Malyshkin V (1988) Parallel program synthesis on the basis of computational models. Nauka, Novosibirsk (In Russian. Sintez Parallel’nykh program I system na Vychislitel’nykh modelyakh)

    Google Scholar 

  15. 15.

    Blumofe RD, Joerg CF, Kuszmaul BC, Leiserson CE, Randall KH, Zhou Y (1995) Cilk: An efficient multithreaded runtime system. ACM SIGPLAN Not 30(8):207–216

    Article  Google Scholar 

  16. 16.

    Foster I, Kesselman C, Tuecke S (1998) Nexus: Runtime support for task-parallel programming languages. Cluster Computing 1(1):95–107

    Article  Google Scholar 

  17. 17.

    Shu W, Kale LV (1991) Chare Kernel – A runtime support system for parallel computations. J Parallel Distrib Comput 11(3):198–211

    Article  Google Scholar 

  18. 18.

    Chien AA, Karamcheti V, Plevyak J (1993) The concert system – compiler and runtime support for efficient, fine-grained concurrent object-oriented programs. UIUC DCS Tech Report R-93-1815

  19. 19.

    Grimshaw AS, Weissman JB, Strayer WT (1996) Portable run-time support for dynamic object-oriented parallel processing. ACM Trans Comput Syst 14(2):139–170

    Article  Google Scholar 

  20. 20.

    Benson GD, Olsson RA (1997) A portable run-time system for the SR concurrent programming language. In: Proceedings of the workshop on run-time systems for parallel programming (RTSPP).

  21. 21.

    Kalgin KV, Malyskin VE, Nechaev SP, Tschukin GA (2007) Runtime system for parallel execution of fragmented subroutines. In: Proceedings of the 9th international conference on parallel computing technologies (PaCT-2007). LNCS, vol 4671. Springer, Berlin, pp 544–552

    Google Scholar 

  22. 22.

    Faddeev DK, Faddeeva VN (1967) Computing methods of linear algebra, 2nd edn. Nauka, Moscow (In Russian. Vychislitel’nye metody lineinoi algebry)

    Google Scholar 

  23. 23.

    Golub GH, Van Loan CF (1996) Matrix computations, 3rd edn. John Hopkins University Press, Baltimore

    MATH  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to S. Kireev.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Kireev, S., Malyshkin, V. Fragmentation of numerical algorithms for parallel subroutines library. J Supercomput 57, 161–171 (2011). https://doi.org/10.1007/s11227-010-0385-3

Download citation

Keywords

  • Asynchronous programming
  • Parallel program
  • Numerical algorithm
  • Fragment-based programming
  • Dynamic programs’ properties