From: Arnaud Giersch <arnaud.giersch@iut-bm.univ-fcomte.fr>
Date: Wed, 13 Mar 2013 15:49:57 +0000 (+0100)
Subject: wip
X-Git-Url: https://bilbo.iut-bm.univ-fcomte.fr/and/gitweb/loba-papers.git/commitdiff_plain/1e4ddc1ac6f6e145312650aa924c3a98871a0dbc?ds=sidebyside

wip
---

diff --git a/supercomp11/supercomp11.tex b/supercomp11/supercomp11.tex
index a7bc09a..2c1cb7c 100644
--- a/supercomp11/supercomp11.tex
+++ b/supercomp11/supercomp11.tex
@@ -638,13 +638,28 @@ With these constraints in mind, we defined the following metrics:
 In this section, the results for the different simulations will be presented,
 and we'll try to explain our observations.
 
-\FIXME{cluster vs. grid : cluster légèrement plus rapide, mais c'est tout -> chose to give results for grid only.}
+\subsubsection{Cluster vs grid platforms}
 
-\FIXME{explain how to read the graphs}
+As mentioned earlier, we simulated the different algorithms on two kinds of
+physical platforms: clusters and grids.  A first observation that we can make,
+is that the graphs we draw from the data have a similar aspect for the two kinds
+of platforms.  The only noticeable difference is that the algorithms need a bit
+more time to achieve the convergence on the grid platforms, than on clusters.
+Nevertheless their relative performances remain generally identical.
 
-\subsubsection{Main results}
+This suggests that the relative performances of the different strategies are not
+influenced by the characteristics of the physical platform.  The differences in
+the convergence times can be explained by the fact that on the grid platforms,
+distant sites are interconnected by links of smaller bandwith.
+
+Therefore, in the following, we'll only discuss the results for the grid
+platforms.  The different results are presented on the
+figures~\ref{fig.results1} and~\ref{fig.resultsN}.
 
-\begin{figure}
+\FIXME{explain how to read the graphs}
+ratio 1:1 not given here
+
+\begin{figure*}[p]
   \centering
   \includegraphics[width=.5\linewidth]{data/graphs/R1-1:10-grid-line}%
   \includegraphics[width=.5\linewidth]{data/graphs/R1-10:1-grid-line}
@@ -654,17 +669,9 @@ and we'll try to explain our observations.
   \includegraphics[width=.5\linewidth]{data/graphs/R1-10:1-grid-hcube}
   \caption{Real mode, initially on an only mode, comp/comm ratio = 1/10 (left), or 10/1 (right).}
   \label{fig.results1}
-\end{figure}
-
-On fig.~\ref{fig.results1}, \dots
-
-\subsubsection{With the virtual load extension}
+\end{figure*}
 
-\subsubsection{The $k$ parameter}
-
-\subsubsection{With an initial random repartition,  and larger platforms}
-
-\begin{figure}
+\begin{figure*}[p]
   \centering
   \includegraphics[width=.5\linewidth]{data/graphs/RN-1:10-grid-line}%
   \includegraphics[width=.5\linewidth]{data/graphs/RN-10:1-grid-line}
@@ -674,7 +681,17 @@ On fig.~\ref{fig.results1}, \dots
   \includegraphics[width=.5\linewidth]{data/graphs/RN-10:1-grid-hcube}
   \caption{Real mode, random initial distribution, comp/comm ratio = 1/10 (left), or 10/1 (right).}
   \label{fig.resultsN}
-\end{figure}
+\end{figure*}
+
+\subsubsection{Main results}
+
+On fig.~\ref{fig.results1}, \dots
+
+\subsubsection{With the virtual load extension}
+
+\subsubsection{The $k$ parameter}
+
+\subsubsection{With an initial random repartition,  and larger platforms}
 
 \subsubsection{With integer load}