<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Hi Ben,<br>
Thanks again for the patches, they made a huge difference, increased
efficiency from 21% to 81%!<br>
<br>
Here are the numbers:<br>
<table x:str="" style="border-collapse: collapse; width: 240pt;"
border="0" cellpadding="0" cellspacing="0" width="320">
<col style="width: 48pt;" span="5" width="64"> <tbody>
<tr style="height: 12.75pt;" height="17">
<td style="height: 12.75pt; width: 48pt;" height="17" width="64"><br>
</td>
<td style="width: 48pt;" width="64">1 Node Perf</td>
<td style="width: 48pt;" width="64">Falkon</td>
<td style="width: 48pt;" width="64">Swift+Falkon</td>
<td style="width: 48pt;" width="64">Swift+Falkon (patched)</td>
</tr>
<tr style="height: 12.75pt;" height="17">
<td style="height: 12.75pt;" height="17">Min</td>
<td x:num="63.618000000000002" align="right">63.618</td>
<td x:num="53.781999999999996" align="right">53.782</td>
<td x:num="169.13900000000001" align="right">169.139</td>
<td x:num="58.537999999999997" align="right">58.538</td>
</tr>
<tr style="height: 12.75pt;" height="17">
<td style="height: 12.75pt;" height="17">Average</td>
<td x:num="" align="right">64.76</td>
<td x:num="65.472528361002603" align="right">65.47253</td>
<td x:num="309.19445117187502" align="right">309.1945</td>
<td x:num="80.212459472656249" align="right">80.21246</td>
</tr>
<tr style="height: 12.75pt;" height="17">
<td style="height: 12.75pt;" height="17">Median</td>
<td x:num="64.740718749999999" align="right">64.74072</td>
<td x:num="64.774000000000001" align="right">64.774</td>
<td x:num="313.55349999999999" align="right">313.5535</td>
<td x:num="76.524500000000003" align="right">76.5245</td>
</tr>
<tr style="height: 12.75pt;" height="17">
<td style="height: 12.75pt;" height="17">Max</td>
<td x:num="65.863" align="right">65.863</td>
<td x:num="94.447000000000003" align="right">94.447</td>
<td x:num="605.654" align="right">605.654</td>
<td x:num="115.23699999999999" align="right">115.237</td>
</tr>
<tr style="height: 12.75pt;" height="17">
<td style="height: 12.75pt;" height="17">Standard Deviation</td>
<td x:num="0.48898368723463459" align="right">0.488984</td>
<td x:num="3.8639440796778195" align="right">3.863944</td>
<td x:num="52.13821160258469" align="right">52.13821</td>
<td x:num="10.956519557081371" align="right">10.95652</td>
</tr>
<tr style="height: 12.75pt;" height="17">
<td style="height: 12.75pt;" height="17">Efficiency</td>
<td class="xl22" x:num="1" x:fmla="=B3/B3" align="right">100%</td>
<td class="xl22" x:num="0.98911713998466877" x:fmla="=B3/C3"
align="right">99%</td>
<td class="xl22" x:num="0.20944748443755615" x:fmla="=B3/D3"
align="right">21%</td>
<td class="xl22" x:num="0.80735586997025244" x:fmla="=B3/E3"
align="right">81%</td>
</tr>
</tbody>
</table>
<br>
The first column shows the per task statistic when running on 1 node (4
CPUs) through Falkon. The second column are the statistics for running
the application at large scale, on 2048 CPUs. The 3rd column is
running Swift+Falkon (both from SVN) on 256 CPUs. The 4th column is
Swift+Falkon, but Swift has the 3 patches applied. Essentially, the
per task execution time was reduced from 309 seconds to 80 seconds,
where the ideal would have been 64 seconds. It brought the efficiency
from 21% to 81% for this particular workload. This looks fantastic! <br>
<br>
We'll have to verify that we can maintain this 81% efficiency to higher
number of CPUs. In the meantime, if you can think of anything else
that we could do to keep pushing the 81% efficiency number higher, let
us know.4<br>
<br>
Thanks again,<br>
Ioan<br>
<br>
Ben Clifford wrote:
<blockquote
cite="mid:Pine.LNX.4.64.0803310723010.9854@dildano.hawaga.org.uk"
type="cite">
<pre wrap="">On Mon, 31 Mar 2008, Ben Clifford wrote:
</pre>
<blockquote type="cite">
<pre wrap="">This temporary directory handling is pretty ugly - it should be a couple
lines change to wrapper.sh to get similar functionality using the existing
swift temporary direcotry handling - change the path to /tmp and use cp
instead of ln -s. That way you can take advantage of Swift's existing
unique job IDs and error handling too.
</pre>
</blockquote>
<pre wrap=""><!---->
Attached are three patches that will apply against svn r1775:
The first puts temporary directories in /tmp rather than on shared fs.
<a class="moz-txt-link-freetext" href="http://www.ci.uchicago.edu/~benc/tmp/wrapper-tmp-dirs-on-tmp">http://www.ci.uchicago.edu/~benc/tmp/wrapper-tmp-dirs-on-tmp</a>
The second copies the application file to the worker in each job execution
(though doesn't do any worker-node caching of such between jobs)
<a class="moz-txt-link-freetext" href="http://www.ci.uchicago.edu/~benc/tmp/wrapper-tmp-dirs-mv-executable">http://www.ci.uchicago.edu/~benc/tmp/wrapper-tmp-dirs-mv-executable</a>
The third creates the worker node log on /tmp and copies it at the end.
<a class="moz-txt-link-freetext" href="http://www.ci.uchicago.edu/~benc/tmp/wrapper-tmp-log-locally">http://www.ci.uchicago.edu/~benc/tmp/wrapper-tmp-log-locally</a>
The three modify all wrapper.sh and should be applied in the above order.
With the first two patches, the timestamps in the usual info logs will
provide information about how long the copies take, in the same way that
they usually indicate times for other execution stages.
</pre>
</blockquote>
<br>
<pre class="moz-signature" cols="72">--
===================================================
Ioan Raicu
Ph.D. Candidate
===================================================
Distributed Systems Laboratory
Computer Science Department
University of Chicago
1100 E. 58th Street, Ryerson Hall
Chicago, IL 60637
===================================================
Email: <a class="moz-txt-link-abbreviated" href="mailto:iraicu@cs.uchicago.edu">iraicu@cs.uchicago.edu</a>
Web: <a class="moz-txt-link-freetext" href="http://www.cs.uchicago.edu/~iraicu">http://www.cs.uchicago.edu/~iraicu</a>
<a class="moz-txt-link-freetext" href="http://dev.globus.org/wiki/Incubator/Falkon">http://dev.globus.org/wiki/Incubator/Falkon</a>
<a class="moz-txt-link-freetext" href="http://dsl-wiki.cs.uchicago.edu/index.php/Main_Page">http://dsl-wiki.cs.uchicago.edu/index.php/Main_Page</a>
===================================================
===================================================
</pre>
</body>
</html>