EasyMSG : tools and techniques for an adaptive overlapping in SPMD programming

Pascal Havé

ESAIM: Mathematical Modelling and Numerical Analysis - Modélisation Mathématique et Analyse Numérique (2002)

  • Volume: 36, Issue: 5, page 863-882
  • ISSN: 0764-583X

Abstract

top
During the development of a parallel solver for Maxwell equations by integral formulations and Fast Multipole Method (FMM), we needed to optimize a critical part including a lot of communications and computations. Generally, many parallel programs need to communicate, but choosing explicitly the way and the instant may decrease the efficiency of the overall program. So, the overlapping of computations and communications may be a way to reduce this drawback. We will see a implementation of this techniques using dynamic and adaptive overlapping based on the EasyMSG high level C++ library over MPI, a case of SPMD programming.

How to cite

top

Havé, Pascal. "EasyMSG : tools and techniques for an adaptive overlapping in SPMD programming." ESAIM: Mathematical Modelling and Numerical Analysis - Modélisation Mathématique et Analyse Numérique 36.5 (2002): 863-882. <http://eudml.org/doc/244927>.

@article{Havé2002,
abstract = {During the development of a parallel solver for Maxwell equations by integral formulations and Fast Multipole Method (FMM), we needed to optimize a critical part including a lot of communications and computations. Generally, many parallel programs need to communicate, but choosing explicitly the way and the instant may decrease the efficiency of the overall program. So, the overlapping of computations and communications may be a way to reduce this drawback. We will see a implementation of this techniques using dynamic and adaptive overlapping based on the EasyMSG high level C++ library over MPI, a case of SPMD programming.},
author = {Havé, Pascal},
journal = {ESAIM: Mathematical Modelling and Numerical Analysis - Modélisation Mathématique et Analyse Numérique},
keywords = {SPMD parallel processing; message passing environment; communications optimization; C++; Maxwell equations; fast multipole method},
language = {eng},
number = {5},
pages = {863-882},
publisher = {EDP-Sciences},
title = {EasyMSG : tools and techniques for an adaptive overlapping in SPMD programming},
url = {http://eudml.org/doc/244927},
volume = {36},
year = {2002},
}

TY - JOUR
AU - Havé, Pascal
TI - EasyMSG : tools and techniques for an adaptive overlapping in SPMD programming
JO - ESAIM: Mathematical Modelling and Numerical Analysis - Modélisation Mathématique et Analyse Numérique
PY - 2002
PB - EDP-Sciences
VL - 36
IS - 5
SP - 863
EP - 882
AB - During the development of a parallel solver for Maxwell equations by integral formulations and Fast Multipole Method (FMM), we needed to optimize a critical part including a lot of communications and computations. Generally, many parallel programs need to communicate, but choosing explicitly the way and the instant may decrease the efficiency of the overall program. So, the overlapping of computations and communications may be a way to reduce this drawback. We will see a implementation of this techniques using dynamic and adaptive overlapping based on the EasyMSG high level C++ library over MPI, a case of SPMD programming.
LA - eng
KW - SPMD parallel processing; message passing environment; communications optimization; C++; Maxwell equations; fast multipole method
UR - http://eudml.org/doc/244927
ER -

References

top
  1. [1] S. Balay, W. Gropp, L. McInnes and B. Smith, Petsc 2.0 users manual. Technical report, Argonne National Laboratory (1996). 
  2. [2] P. Havé, A parallel implementation of the Fast Multipole Method for Maxwell equations. Number Eccomas2001-7. Laboratoire d’Analyse Numérique de l’Université Pierre et Marie Curie, John Wiley & Sons (2001). Zbl1037.78021
  3. [3] J.C. Nédélec, Cours de DEA de l’École Polytechnique et de l’Université Paris 6 (1999). 
  4. [4] S.M. Rao, D.R. Wilton and A.W. Glisson, Electromagnetic scattering by surfaces of arbitrary shape. IEEE Trans. Antennas and Propagations 30 (1982) 409–418. 
  5. [5] J.V.W. Reynders, The POOMA FrameWork—a templated class library for parallel scientific computing, in Proceedings of the Eighth SIAM Conference on Parallel Processing for Scientific Computing (Minneapolis, MN, 1997), Philadelphia, PA (1997) SIAM, p. 6. MR1602540
  6. [6] V. Rokhlin, Rapid solution of integral equations of scattering theory in two dimensions. J. Comput. Phys. 86 (1990) 414–439. Zbl0686.65079
  7. [7] Y. Roudier, D. Caromel and F. Belloncle, The C++// System, in Parallel Programming Using C++, G. Wilson and P. Lu Eds., MIT Press (1996) 257–296. 
  8. [8] D. Sagnol, F. Baude, D. Caromel and N. Furmento, Overlapping communication with computation in distributed object systems. Lecture Notes Comput. Sci. 1593, Springer, Amsterdam (1999) 744–753. Zbl0997.68619
  9. [9] D.C. Schmidt and T. Suda, Measuring the performance of parallel message-based process architectures. INFOCOM 2 (1995) 624–633. 
  10. [10] A. Skjellum, W. Gropp and E. Lusk, Using MPI: portable parallel programming with the message passing interface. ISBN 0-262-57104-8. MIT Press (1994). Zbl0875.68206
  11. [11] J. Vayssiere, D. Caromel and W. Klauser, Towards seamless computing and metacomputing in Java, in Concurrency Practice and Experience, G.C. Fox Ed., Wiley & Sons, Ltd (1998) 1043–1061. 

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.