\relax 
\@writefile{toc}{\author{Lilia Ziane Khodja}{}}
\@writefile{toc}{\author{Rapha\IeC {\"e}l Couturier}{}}
\@writefile{toc}{\author{Jacques Bahi}{}}
\@writefile{loa}{\addvspace {10\p@ }}
\@writefile{toc}{\contentsline {chapter}{\numberline {13}Solving sparse linear systems with GMRES and CG methods on GPU clusters}{313}}
\@writefile{lof}{\addvspace {10\p@ }}
\@writefile{lot}{\addvspace {10\p@ }}
\newlabel{ch12}{{13}{313}}
\@writefile{toc}{\contentsline {section}{\numberline {13.1}Introduction}{313}}
\newlabel{ch12:sec:01}{{13.1}{313}}
\@writefile{toc}{\contentsline {section}{\numberline {13.2}Krylov iterative methods}{314}}
\newlabel{ch12:sec:02}{{13.2}{314}}
\newlabel{ch12:eq:01}{{13.1}{314}}
\newlabel{ch12:eq:02}{{13.2}{314}}
\newlabel{ch12:eq:03}{{13.3}{314}}
\newlabel{ch12:eq:11}{{13.4}{315}}
\@writefile{toc}{\contentsline {subsection}{\numberline {13.2.1}CG method}{315}}
\newlabel{ch12:sec:02.01}{{13.2.1}{315}}
\newlabel{ch12:eq:04}{{13.5}{315}}
\newlabel{ch12:eq:05}{{13.6}{315}}
\newlabel{ch12:eq:06}{{13.7}{315}}
\newlabel{ch12:eq:07}{{13.8}{315}}
\newlabel{ch12:eq:08}{{13.9}{315}}
\newlabel{ch12:eq:09}{{13.10}{315}}
\@writefile{loa}{\contentsline {algocf}{\numberline {15}{\ignorespaces Left-preconditioned CG method\relax }}{316}}
\newlabel{ch12:alg:01}{{15}{316}}
\newlabel{ch12:eq:10}{{13.11}{316}}
\@writefile{toc}{\contentsline {subsection}{\numberline {13.2.2}GMRES method}{317}}
\newlabel{ch12:sec:02.02}{{13.2.2}{317}}
\newlabel{ch12:eq:12}{{13.12}{317}}
\newlabel{ch12:eq:13}{{13.13}{317}}
\newlabel{ch12:eq:14}{{13.14}{317}}
\newlabel{ch12:eq:15}{{13.15}{317}}
\newlabel{ch12:eq:16}{{13.16}{317}}
\newlabel{ch12:eq:17}{{13.17}{317}}
\newlabel{ch12:eq:18}{{13.18}{317}}
\newlabel{ch12:eq:19}{{13.19}{317}}
\@writefile{loa}{\contentsline {algocf}{\numberline {16}{\ignorespaces Left-preconditioned GMRES method with restarts\relax }}{318}}
\newlabel{ch12:alg:02}{{16}{318}}
\@writefile{toc}{\contentsline {section}{\numberline {13.3}Parallel implementation on a GPU cluster}{319}}
\newlabel{ch12:sec:03}{{13.3}{319}}
\@writefile{toc}{\contentsline {subsection}{\numberline {13.3.1}Data partitioning}{319}}
\newlabel{ch12:sec:03.01}{{13.3.1}{319}}
\@writefile{lof}{\contentsline {figure}{\numberline {13.1}{\ignorespaces A data partitioning of the sparse matrix $A$, the solution vector $x$ and the right-hand side $b$ into four portions.\relax }}{320}}
\newlabel{ch12:fig:01}{{13.1}{320}}
\@writefile{toc}{\contentsline {subsection}{\numberline {13.3.2}GPU computing}{320}}
\newlabel{ch12:sec:03.02}{{13.3.2}{320}}
\@writefile{toc}{\contentsline {subsection}{\numberline {13.3.3}Data communications}{321}}
\newlabel{ch12:sec:03.03}{{13.3.3}{321}}
\@writefile{lof}{\contentsline {figure}{\numberline {13.2}{\ignorespaces Data exchanges between \textit  {Node 1} and its neighbors \textit  {Node 0}, \textit  {Node 2} and \textit  {Node 3}.\relax }}{322}}
\newlabel{ch12:fig:02}{{13.2}{322}}
\@writefile{lof}{\contentsline {figure}{\numberline {13.3}{\ignorespaces Columns reordering of a sparse sub-matrix.\relax }}{323}}
\newlabel{ch12:fig:03}{{13.3}{323}}
\@writefile{toc}{\contentsline {section}{\numberline {13.4}Experimental results}{324}}
\newlabel{ch12:sec:04}{{13.4}{324}}
\@writefile{lof}{\contentsline {figure}{\numberline {13.4}{\ignorespaces General scheme of the GPU cluster of tests composed of six machines, each with two GPUs.\relax }}{324}}
\newlabel{ch12:fig:04}{{13.4}{324}}
\@writefile{lof}{\contentsline {figure}{\numberline {13.5}{\ignorespaces Sketches of sparse matrices chosen from the Davis collection.\relax }}{325}}
\newlabel{ch12:fig:05}{{13.5}{325}}
\@writefile{lot}{\contentsline {table}{\numberline {13.1}{\ignorespaces Main characteristics of sparse matrices chosen from the Davis collection.\relax }}{325}}
\newlabel{ch12:tab:01}{{13.1}{325}}
\@writefile{lot}{\contentsline {table}{\numberline {13.2}{\ignorespaces Performances of the parallel CG method on a cluster of 24 CPU cores vs. on a cluster of 12 GPUs.\relax }}{326}}
\newlabel{ch12:tab:02}{{13.2}{326}}
\@writefile{lot}{\contentsline {table}{\numberline {13.3}{\ignorespaces Performances of the parallel GMRES method on a cluster 24 CPU cores vs. on cluster of 12 GPUs.\relax }}{326}}
\newlabel{ch12:tab:03}{{13.3}{326}}
\newlabel{ch12:eq:20}{{13.20}{327}}
\@writefile{lof}{\contentsline {figure}{\numberline {13.6}{\ignorespaces Parallel generation of a large sparse matrix by four computing nodes.\relax }}{328}}
\newlabel{ch12:fig:06}{{13.6}{328}}
\@writefile{lot}{\contentsline {table}{\numberline {13.4}{\ignorespaces Main characteristics of sparse banded matrices generated from those of the Davis collection.\relax }}{328}}
\newlabel{ch12:tab:04}{{13.4}{328}}
\@writefile{lot}{\contentsline {table}{\numberline {13.5}{\ignorespaces Performances of the parallel CG method for solving linear systems associated to sparse banded matrices on a cluster of 24 CPU cores vs. on a cluster of 12 GPUs.\relax }}{329}}
\newlabel{ch12:tab:05}{{13.5}{329}}
\@writefile{toc}{\contentsline {section}{\numberline {13.5}Conclusion}{329}}
\newlabel{ch12:sec:05}{{13.5}{329}}
\@writefile{lot}{\contentsline {table}{\numberline {13.6}{\ignorespaces Performances of the parallel GMRES method for solving linear systems associated to sparse banded matrices on a cluster of 24 CPU cores vs. on a cluster of 12 GPUs.\relax }}{330}}
\newlabel{ch12:tab:06}{{13.6}{330}}
\@writefile{toc}{\contentsline {section}{Bibliography}{330}}
\@setckpt{Chapters/chapter12/ch12}{
\setcounter{page}{332}
\setcounter{equation}{22}
\setcounter{enumi}{2}
\setcounter{enumii}{0}
\setcounter{enumiii}{0}
\setcounter{enumiv}{10}
\setcounter{footnote}{0}
\setcounter{mpfootnote}{0}
\setcounter{part}{5}
\setcounter{chapter}{13}
\setcounter{section}{5}
\setcounter{subsection}{0}
\setcounter{subsubsection}{0}
\setcounter{paragraph}{0}
\setcounter{subparagraph}{0}
\setcounter{figure}{6}
\setcounter{table}{6}
\setcounter{numauthors}{0}
\setcounter{parentequation}{46}
\setcounter{subfigure}{0}
\setcounter{lofdepth}{1}
\setcounter{subtable}{0}
\setcounter{lotdepth}{1}
\setcounter{lstnumber}{50}
\setcounter{ContinuedFloat}{0}
\setcounter{AlgoLine}{29}
\setcounter{algocfline}{16}
\setcounter{algocfproc}{16}
\setcounter{algocf}{16}
\setcounter{nprt@mantissa@digitsbefore}{0}
\setcounter{nprt@mantissa@digitsafter}{0}
\setcounter{nprt@exponent@digitsbefore}{0}
\setcounter{nprt@exponent@digitsafter}{0}
\setcounter{nprt@digitsfirstblock}{0}
\setcounter{nprt@blockcnt}{0}
\setcounter{nprt@cntprint}{0}
\setcounter{proposition}{1}
\setcounter{theorem}{0}
\setcounter{exercise}{0}
\setcounter{example}{0}
\setcounter{definition}{0}
\setcounter{proof}{1}
\setcounter{lstlisting}{0}
}