Author: David Schneider <[email protected]>
Branch: extradoc
Changeset: r4783:e2d42d76e927
Date: 2012-09-07 11:59 +0200
http://bitbucket.org/pypy/extradoc/changeset/e2d42d76e927/
Log: Add some notes about the changes to the tables
diff --git a/talk/vmil2012/paper.tex b/talk/vmil2012/paper.tex
--- a/talk/vmil2012/paper.tex
+++ b/talk/vmil2012/paper.tex
@@ -728,7 +728,8 @@
\label{fig:benchmarks}
\end{figure*}
-Figure~\ref{fig:benchmarks} summarizes the total number of operations that were
+Figure~\ref{fig:benchmarks} summarizes\footnote{In all tables the minimum and
+maximum values for each column are highlighted in dark/light gray.} the total
number of operations that were
recorded during tracing for each of the benchmarks and what percentage of these
operations are guards. The static number of operations was counted on the
unoptimized
and optimized traces. The figure also shows the overall optimization rate for
@@ -749,7 +750,8 @@
The threshold used is 200 failures. This rather high threshold was picked
experimentally to give
good results for long-running programs.
}
-
+It also contains sparklines depicting the failure rates for the failing guards
+in decreasing order, each normalized to the most failing guard.
The numbers presented for guards that have a bridge represent the
failures up to the compilation of the bridge and all executions of the then
attached bridge.
@@ -768,7 +770,9 @@
Also, of all failing guards a few fail extremely often
and most fail rarely. Reinforcing this notion the figure shows that, depending
on the
benchmark, between 0.008\% and 0.225\% of the guards are responsible for 50\%
-of the total guards failures.
+of the total guards failures. Even considering 99.9\% of guard failures the
+relative amount of guards does not rise above 3\%.
+The colored dots in the sparklines correspond to 50\%, 99\% and 99.9\%.
These results emphasize that as most of the guards never
fail it is important to make sure that the successful execution of a guard does
not have unnecessary overhead.
@@ -1029,5 +1033,5 @@
%\end{figure*}
\bibliographystyle{abbrv}
\bibliography{zotero,paper}
-\listoftodos
+%\listoftodos
\end{document}
_______________________________________________
pypy-commit mailing list
[email protected]
http://mail.python.org/mailman/listinfo/pypy-commit
