<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
The X5675 supports hyperthreading. Does i.e. "mpiexec -n 2 maker"
use 2 physical cores or 2 threads of the 1st core? If the latter
happens it would be interesting to see your series extended to -n
24.<br>
<br>
Cheers<br>
Olaf<br>
<br>
<div class="moz-cite-prefix">On 03/14/2013 11:19 AM, Ramón Fallon
wrote:<br>
</div>
<blockquote
cite="mid:CA+WKpEtg+j+wLp_qntSWff1RMXyVS5hRzz=fc+Amk6OMmqdfPA@mail.gmail.com"
type="cite">Hi,
<div><br>
</div>
<div>I was trying to tweak some of our machines to maximise
Mpich2/Maker (svn rev 997) throughput and describe one small
set of results on this mailing list to allow sharing of
experiences.</div>
<div><br>
</div>
<div>I use the example input dataset "dpp_contig.fasta" with the
original sequence repeated 125 times within the same file (under
different names of course) to allow for a decent size run. This
file totalled 4.019 megabases. I use the dpp_proteins.fasta and
The maker_opts.ctl has "cpus=1" set as the docs recommend for
MPI.</div>
<div><br>
</div>
<div>Hardware is a standalone HP Proliant SL390 with two
Intel X5675 @ 3.07GHz, totalling 12 cores with 192GB RAM and 1TB
disk (local, no NFS) running Ubuntu 10.04 with 2.6.32-41 linux
kernel </div>
<div><br>
</div>
<div>commandline was "mpiexec -n <#cores> maker" within a
dedicated directory containing all relevant files.</div>
<div><br>
</div>
<div>
<div> #cores time(mins) Megabases/hr</div>
<div> 1 27.00 8.93</div>
<div> 2 126.25 1.91</div>
<div> 4 42.57 5.66</div>
<div> 6 25.42 9.49</div>
<div> 8 18.60 12.96</div>
<div> 10 16.67 14.47</div>
<div> 12 13.98 17.24</div>
</div>
<div><br>
</div>
<div>I attach a png file with graph. The upshot of this particular
experiment is that 2 processes show anomalous behaviour and that
6 processors are needed to gain an advantage on the 1 processor
run, while 12 processors achieves a speed-up of nearly 2 on the
1 processor version.</div>
<div><br>
</div>
<div>I am now going to move on to a three node cluster with 2x
8core processors each (so I can go up to 48 processors), so will
report back with higher core numbers. Any suggestions on further
speed optimizations welcome.</div>
<div><br>
</div>
<div>Cheers / Ramón.</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
<pre wrap="">_______________________________________________
maker-devel mailing list
<a class="moz-txt-link-abbreviated" href="mailto:maker-devel@box290.bluehost.com">maker-devel@box290.bluehost.com</a>
<a class="moz-txt-link-freetext" href="http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org">http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org</a>
</pre>
</blockquote>
<br>
<pre class="moz-signature" cols="72">
</pre>
</body>
</html>