Object Oriented Programming in High Performance Fortran

  • E. de Sturler
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1543)

Abstract

High Performance Fortran[4] defines extensions for parallel programming to the Fortran standard; with the definition of Fortran 95, this means mainly data distribution and alignment and several intrinsic and library functions. We have shown that High Performance Fortran can be used to write very efficient parallel programs for regular and even irregular matrix problems [1,2] (see http://www.scsc.ethz.ch/homes/Vonsturler). However, complex sparse matrix data structures and parallel adaptations of these lead to rather complex programs.

Keywords

High Performance Fortran Object-oriented Programming DataDistribution Regular Sparse Matrices Irregular Sparse Matrices Iterative Methods 

References

  1. 1.
    E. De Sturler and D. Loher. Implementing iterative solvers for irregular sparse matrix problems in High Performance Fortran. In C. Polychronopoulos, K. Joe, K. Araki, and M. Amamiya, editors, International Symposium on High Performance Computing, ISHPC’97, Lecture Notes in Computer Science 1336, pages 293–304, Berlin, Heidelberg, Germany, 1997. Springer-Verlag. Proceedings of ISHPC’97, Fukuoka, Japan, November 1997.CrossRefGoogle Scholar
  2. 2.
    E. De Sturler and D. Loher. Parallel iterative solvers for irregular sparse matrices in High Performance Fortran. Future Generation Computer Systems, 13:315–325, 1998. draft: see http://www.scsc.ethz.ch/sturler/Public/fgcs97f.ps.gz.CrossRefGoogle Scholar
  3. 3.
    High Performance Fortran Forum. High Performance Fortran Language Specification, version 1.1. Rice University, 1994.Google Scholar
  4. 4.
    High Performance Fortran Forum. High Performance Fortran Language Specification, version 2.0. Rice University, 1997.Google Scholar
  5. 5.
    C.D. Norton, B.K. Szymanski, and V.K. Decyk. Object oriented parallel computation for plasma simulation. Comm. ACM, 38:88–100, 1995.CrossRefGoogle Scholar
  6. 6.
    C.D. Norton. Object Oriented Programming Paradigms in Scientific Computing. PhD thesis, Rensselaer Polytechnic Institute, Troy, New York, August1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • E. de Sturler
    • 1
  1. 1.Swiss Center for Scientific Computing (SCSC-ETHZ)Swiss Federal Institute of Technology Zurich, ETH Zentrum (RZ F-11)ZurichSwitzerland

Personalised recommendations