Fabrice RASTELLO  Home/Search
Context Related
| perso.enslyon.fr/fab...RRLIP200142.ps Cached: PS.gz PS PDF This document uses CoBlitz to cache paper downloads. If your firewall is blocking outgoing connections to port 3125, you can use these links to download local copies.
Image Update HelpPS.gz PS PDF From: perso.enslyon.fr/fabri...reports (more) (Enter author homepages) |
Rate this article:      (best)Comment on this article |
Abstract: In this paper, we present an ecient and simple solution to the parallelization of discrete integration programs of ordinary di erential equations (ODE). The main technique used is known as loop tiling. To avoid the overhead due to code complexity and border e ects, we introduce redundant tasks and we use non parallelepiped tiles. Thanks both to cache reuse (4:3) and coarse granularity (24:5) , the speedup using 25 processors over the non-tiled sequential implementation is larger than... (Update)
Cited by: More
Efficient tiling for an ODE discrete integration program: .. - Instead Of Trapezoidal (Correct)
Active bibliography (related documents): More All
0.5: Dynamical Ensembles in Nonequilibrium Statistical Mechanics.. - Rondoni, Stöcker (Correct)
0.3: Optimal Semi-Oblique Tiling - Andonov, Balev, Rajopadhye (2001) (Correct)
0.3: Time-Minimal Tiling When Rise is Larger Than Zero - Xue, Cai (2002) (Correct)
Similar documents based on text: More All
0.4: Lower Bounds Are not Easier over the Reals: Inside PH - Fournier, Koiran (1999) (Correct)
0.4: Workshop "Compilation et Parallélisation Automatique" - Mongenet, Rajopadhye.. (1999) (Correct)
0.4: Further Reducing the Redundancy of a Notation Over a.. - Daumas, Matula (2000) (Correct)
BibTeX entry: (Update)
F. Rastello and T. Dauxois. Parallelization of the numerical lyapunov calculation for the fermi-pasta-ulam chain. Technical Report RR-01-42, LIP, ENS Lyon, France, November 2001. Available at www.ens-lyon.fr/LIP/. http://citeseer.ist.psu.edu/rastello01parallelization.html More
@misc{ rastello01parallelization,
author = "F. Rastello and T. Dauxois",
title = "Parallelization of the numerical lyapunov calculation for the fermi-pasta-ulam
chain",
text = "F. Rastello and T. Dauxois. Parallelization of the numerical lyapunov calculation
for the fermi-pasta-ulam chain. Technical Report RR-01-42, LIP, ENS Lyon,
France, November 2001. Available at www.ens-lyon.fr/LIP/.",
year = "2001",
url = "citeseer.ist.psu.edu/rastello01parallelization.html" }
Citations (may not include all citations):294 A loop transformation theory and an algorithm to maximize pa.. (context) - Wolf, Lam - 1991
126 BIP: a new protocol designed for high performance networking.. (context) - Prylli, Tourancheau - 1998
67 Maximizing parallelism and minimizing synchronization with a.. - Lim, Lam - 1997
36 An introduction to a formal theory of dependence analysis (context) - Banerjee - 1988
35 Combining retiming and scheduling techniques for loop parall.. - Darte, Silber et al. - 1997
34 and Yves Robert (context) - Boulet, Darte et al. - 1994
32 Regular and Chaotic Dynamics (context) - Lichtenberg, Lieberman - 1992
25 Deterministic nonperiodic ow (context) - Lorenz - 1963
21 Hierarchical tiling: a methodology for high performance - Carter, Ferrante et al. - 1996
20 A proposal for a heterogeneous cluster ScaLAPACK (context) - Boudet, Rastello et al. - 1999
19 Lyapunov characteristic exponents for smooth dynamical syste.. (context) - Benettin - 1980
19 Determining the idle time of a tiling (context) - ogstedt, Carter et al. - 1997
15 Algorithmic issues on heterogeneous computing platforms - Boulet, Dongarra et al. - 1999
15 Tiling with limited resources - Calland, Dongarra et al. - 1997
12 The accuracy of symplectic integrators - MacLachlan, Atela - 1992
12 Precise tiling for uniform loop nests - Calland, Risset - 1995
11 Optimal orthogonal tiling (context) - Andonov, Rajopadhye et al. - 1998
6 Determining the idle time of a tiling: new results - Desprez, Dongarra et al. - 1998
5 Modeling of a high speed network to maximize throughput perf.. - Prylli, Tourancheau et al.
3 Relaxation properties and ergodicity breaking in nonlinear h.. (context) - Pettini, Landol - 1990
3 Optimal task scheduling to minimize inter-tile latencies - Rastello, Rao et al. - 1998
3 heterogeneous computing: A prospective report (context) - Mehaut, Robert et al. - 1999
2 The number dependence of the maximum lyapunov exponent (context) - Searles, Evans et al. - 1997
2 Automatic partitioning of parallel loops with parallelepiped.. - Rastello, Robert - 2001
1 Science laboratory report no (context) - Fermi, Pasta et al. - 1965
1 and John Mellor-Crummey (context) - Chavarra-Miranda, Darte et al. - 2001
1 Distribution of characteristic exponents in the thermodynami.. (context) - Livi, Politi et al. - 1986
1 Localization and equipartition of energy in the -fpu chain :.. (context) - Cretegny, Dauxois et al. - 1998
1 Gaussian model for chaotic instability of hamiltonian ows - Casetti, Livi et al. - 1996
1 Modulational estimate for the maximal lyapunov exponent in f.. (context) - Dauxois, Ru et al. - 1997
Documents on the same site (http://perso.ens-lyon.fr/fabrice.rastello/Biblio/Keyword/../Category/reports.html): More
Dense Linear Algebra Kernels on Heterogeneous.. - Beaumont, Legrand, .. (2000) (Correct)
Laboratoire de l'Informatique du Parall elisme - Ecole Normale Superieure (2001) (Correct)
Optimizing the translation out-of-SSA with renaming.. - Rastello, de Ferriere.. (2003) (Correct)
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC