From: raphael couturier Date: Sun, 27 Apr 2014 10:20:57 +0000 (+0200) Subject: coquille X-Git-Tag: hpcc2014_submission~57^2~4 X-Git-Url: https://bilbo.iut-bm.univ-fcomte.fr/and/gitweb/hpcc2014.git/commitdiff_plain/201b9e3feb0d4318510a67c834f9645f04c4e8d0?ds=sidebyside;hp=-c coquille --- 201b9e3feb0d4318510a67c834f9645f04c4e8d0 diff --git a/hpcc.tex b/hpcc.tex index d60fd6a..dabef52 100644 --- a/hpcc.tex +++ b/hpcc.tex @@ -45,7 +45,7 @@ \begin{document} -\title{Simulation of Asynchronous Iterative Numerical Algorithms Using SimGrid} +\title{Simulation of Asynchronous Iterative Algorithms Using SimGrid} \author{% \IEEEauthorblockN{% @@ -73,9 +73,9 @@ \begin{abstract} -Synchronous iterative algorithms is often less scalable than asynchronous +Synchronous iterative algorithms are often less scalable than asynchronous iterative ones. Performing large scale experiments with different kind of -networks parameters is not easy because with supercomputers such parameters are +network parameters is not easy because with supercomputers such parameters are fixed. So one solution consists in using simulations first in order to analyze what parameters could influence or not the behaviors of an algorithm. In this paper, we show that it is interesting to use SimGrid to simulate the behaviors @@ -83,7 +83,7 @@ of asynchronous iterative algorithms. For that, we compare the behaviour of a synchronous GMRES algorithm with an asynchronous multisplitting one with simulations in which we choose some parameters. Both codes are real MPI codes. Experiments allow us to see when the multisplitting algorithm can be more -efficience than the GMRES one to solve a 3D Poisson problem. +efficient than the GMRES one to solve a 3D Poisson problem. % no keywords for IEEE conferences