High Performance Multigrid on Current Large Scale Parallel Computers

Zusammenfassung
Making multigrid algorithms run efficiently on large parallel computers is a challenge. Without clever data structures the communication overhead will lead to an unacceptable performance drop when using thousands of processors. We show that with a good implementation it is possible to solve a linear system with 1011 unknowns in about 1.5 minutes on almost 10,000 processors. The data structures also allow for efficient adaptive mesh refinement, opening a wide range of applications to our solver.
  • Vollständige Referenz
  • BibTeX
Gradl, T. & Rüde, U., (2008). High Performance Multigrid on Current Large Scale Parallel Computers. In: Nagel, W. E., Hoffmann, R. & Koch, A. (Hrsg.), 9th workshop on parallel systems and algorithms – workshop of the GI/ITG special interest groups PARS and PARVA. Bonn: Gesellschaft für Informatik e. V.. (S. 37-45).
@inproceedings{mci/Gradl2008,
author = {Gradl, Tobias AND Rüde, Ulrich},
title = {High Performance Multigrid on Current Large Scale Parallel Computers},
booktitle = {9th workshop on parallel systems and algorithms – workshop of the GI/ITG special interest groups PARS and PARVA},
year = {2008},
editor = {Nagel, Wolfgang E. AND Hoffmann, Rolf AND Koch, Andreas} ,
pages = { 37-45 },
publisher = {Gesellschaft für Informatik e. V.},
address = {Bonn}
}
DateienGroesseFormatAnzeige
37.pdf138.5Kb PDF Öffnen

Haben Sie fehlerhafte Angaben entdeckt? Sagen Sie uns Bescheid: Feedback abschicken

Mehr Information

ISBN: 978-3-88579-218-5
ISSN: 1617-5468
Datum: 2008
Sprache: en (en)
Typ: Text/Conference Paper
Sammlungen

Zur Langanzeige