<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>Hi Berk:<div><br></div><div>Thank you for your reply. I am doing the visualization on a distributed cluster. And the StreamTracer filter is used with a point source of radius 0.1 and 100 seeds. </div><div><br></div><div>It is surprising that the streamtracer in parallel takes orders of magnitude more time than when applied in serial. I tried to first apply D3 filter and then the streamtracer. That helps a little bit but still the time is prohibitively long (about an hour compared to ~5min when done in serial). I do not know if I am missing something or this is usual.</div><div><br></div><div>Here is the visualization that I want to create in parallel (I had to do it in serial because of this problem):</div><div>https://www.dropbox.com/s/r3kkon3khjdpqs6/foam_3.png</div><div><br></div><div>The original dataset is about 4GB. Here is a smaller one (~1GB) with the state files: </div><div>https://www.dropbox.com/s/r6lh3sx2wojfva2/sample.tar.gz </div><div><br></div><div><div>I would highly appreciate any suggestions.</div><div>-Amir</div><br><div><hr id="stopSpelling">From: berk.geveci@kitware.com<br>Date: Mon, 21 Apr 2014 11:34:25 -0400<br>To: andy.bauer@kitware.com<br>CC: paraview@paraview.org<br>Subject: Re: [Paraview] Fwd: Stream Tracer Scaling<br><br><div dir="ltr">Actually, if the integration is fairly load balanced across partitions, the streamline integration can scale OK. It is highly dependent on how the seeds are distributed and if the integration ends up clustering on a few processes. Amir:<div>
<br></div><div>- How big is your seed source? Does it span multiple partitions? How many seed points?</div><div>- Do the streamlines stay together or do the move around the domain more independently?</div><div><br></div>
<div>
Are you running on a single machine by the way? I have been thinking about parallelizing the streamline filter over threads, which should scale better on a single machine.</div><div><br></div><div>Best,</div><div>-berk</div>
</div><div class="ecxgmail_extra"><br></div>
<br>_______________________________________________
Powered by www.kitware.com
Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html
Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView
Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview</div></div><div class="vimiumReset vimiumHUD" style="right: 150px; opacity: 0; display: none;"></div> </div></body>
</html>