<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
--></style>
</head>
<body class='hmmessage'>
<BR>
Dear all,<BR>
<BR>
so based on the timer log results that i collected, what i read about offscreen-rendering and the explanation below, can i conclude the followings? someone pls correct me if i'm wrong!<BR>
<BR>
with --use-offscreen-rendering<BR>
<UL>
<LI>X windows are disabled on all server processes</LI>
<LI>all server processes render the geometry locally</LI>
<LI>then each server processes sends the image to the client for compositing</LI>
<LI>there's NO data gathering to process 0, meaning the master doesn't collect all images from each processes for composting, then only send it to the client for rendering</LI>
<LI>that's why --use-offscreen-rendering is faster</LI></UL>
<BR>
<BR>
without --use-offscreen-rendering<BR>
<UL>
<LI>X windows are enabled on all server processes</LI>
<LI>all server processes the raw data, then send all geometry to process 0</LI>
<LI>process 0 gather all geometry then send them to the client to render</LI>
<LI>the client does all the rendering job</LI>
<LI>that's why without --use-offscreen-rendering is slower</LI></UL>
<BR>
<BR>
any help / reply is highly appreciated<BR> <BR>Best Regards,<BR>chewping<BR>
<BR> <BR>
<HR id=stopSpelling>
Date: Fri, 20 Nov 2009 10:57:25 +0800<BR>From: weaponfire2005@163.com<BR>To: lcp81um@msn.com<BR>CC: paraview@paraview.org<BR>Subject: Re:[Paraview] Off Screen Rendering<BR><BR>
<DIV>hi chew ping：</DIV>
<DIV> The Problem:"Display is not accessible on the server side. Remote rendering will be disabled.", it means that not all server start up X Server, so the server will process the raw data and send all geometry to the client to render.</DIV>
<DIV> The client do all the rendering job, that's the reason why case 2 is so slow, I think.</DIV>
<DIV> When you use off screen, all server processes render the geometry locally , send image to client to composite, so case 1 much faster than case 2.</DIV>
<DIV> I hope this can help you, for I got the same problem before. Good luck.</DIV>
<DIV><BR><BR><BR> </DIV>
<DIV></DIV><BR>在2009-11-20，"chew ping" <lcp81um@msn.com> 写道：<BR>
<BLOCKQUOTE style="BORDER-LEFT: #ccc 1px solid; PADDING-LEFT: 1ex" id=ecxisReplyContent>Dear all,<BR> <BR>i'm doing parallel rendering using 2 machines (np4), i notice an obvious difference between using offscreen-rendering (faster) and without offscreen-rendering (much slower). i realize this from the timer log:<BR> <BR> <BR><STRONG>Case 1: with offscreen-rendering</STRONG><BR>----------------------------------------------------------<BR><SPAN lang=EN> <BR>Local Process<BR>Still Render, 0.666444 seconds<BR>Execute vtkMPIMoveData id: 519, 0.000199 seconds<BR>Execute vtkPolyDataMapper id: 311, 0.000106 seconds<BR> <BR> <BR>Server, Process 0<BR>Execute vtkFileSeriesReader id: 238, 0.645881 seconds<BR>Execute vtkPVGeometryFilter id: 305, 0.005891 seconds<BR>Execute vtkPVCacheKeeper id: 516, 7.8e-05 seconds<BR>Execute vtkMPIMoveData id: 519, 0.000212 seconds<BR>Execute vtkOrderedCompositeDistributor , 0.000152 seconds<BR>Execute vtkPolyDataMapper id: 311, 0.000274 seconds<BR> <BR>Server, Process 1<BR>Execute vtkFileSeriesReader id: 238, 0.000275 seconds<BR>Execute vtkPVGeometryFilter id: 305, 0.005299 seconds<BR>Execute vtkPVCacheKeeper id: 516, 7.7e-05 seconds<BR>Execute vtkMPIMoveData id: 519, 0.000147 seconds<BR>Execute vtkOrderedCompositeDistributor , 0.000117 seconds<BR>Execute vtkPolyDataMapper id: 311, 9.9e-05 seconds<BR> <BR>Server, Process 2<BR>Execute vtkFileSeriesReader id: 238, 0.000258 seconds<BR>Execute vtkPVGeometryFilter id: 305, 0.004765 seconds<BR>Execute vtkPVCacheKeeper id: 516, 7.7e-05 seconds<BR>Execute vtkMPIMoveData id: 519, 0.000147 seconds<BR>Execute vtkOrderedCompositeDistributor , 0.000111 seconds<BR>Execute vtkPolyDataMapper id: 311, 9.5e-05 seconds<BR> <BR>Server, Process 3<BR>Execute vtkFileSeriesReader id: 238, 0.000352 seconds<BR>Execute vtkPVGeometryFilter id: 305, 0.005351 seconds<BR>Execute vtkPVCacheKeeper id: 516, 7.7e-05 seconds<BR>Execute vtkMPIMoveData id: 519, 0.000168 seconds<BR>Execute vtkOrderedCompositeDistributor , 0.000111 seconds<BR>Execute vtkPolyDataMapper id: 311, 0.000108 seconds<BR> <BR>---------------------------------------------------------------<BR> <BR> <BR> <BR><STRONG>Case 2: without offscreen-rendering</STRONG><BR>---------------------------------------------------------------<BR> <BR><SPAN lang=EN>Local Process<BR>Still Render, 4.86495 seconds<BR>Execute vtkMPIMoveData id: 520, 2.81566 seconds<BR>Execute vtkPolyDataMapper id: 312, 0.00012 seconds<BR>Execute vtkPolyDataMapper id: 148, 7.6e-05 seconds <BR> <BR> <BR>Server, Process 0<BR>Execute vtkFileSeriesReader id: 239, 0.659133 seconds<BR>Execute vtkPVGeometryFilter id: 306, 0.026125 seconds<BR>Execute vtkPVCacheKeeper id: 517, 7.8e-05 seconds<BR>Execute vtkMPIMoveData id: 520, 2.53214 seconds<BR>Dataserver gathering to 0, 1.98166 seconds<BR>Dataserver sending to client, 0.549715 seconds<BR>Execute vtkOrderedCompositeDistributor , 0.000175 seconds<BR> <BR>Server, Process 1<BR>Execute vtkFileSeriesReader id: 239, 0.000446 seconds<BR>Execute vtkPVGeometryFilter id: 306, 0.026313 seconds<BR>Execute vtkPVCacheKeeper id: 517, 7.3e-05 seconds<BR>Execute vtkMPIMoveData id: 520, 0.232435 seconds<BR>Dataserver gathering to 0, 0.232117 seconds<BR>Execute vtkOrderedCompositeDistributor , 0.00012 seconds<BR> <BR>Server, Process 2<BR>Execute vtkFileSeriesReader id: 239, 0.000519 seconds<BR>Execute vtkPVGeometryFilter id: 306, 0.026063 seconds<BR>Execute vtkPVCacheKeeper id: 517, 7e-05 seconds<BR>Execute vtkMPIMoveData id: 520, 1.31373 seconds<BR>Dataserver gathering to 0, 1.31341 seconds<BR>Execute vtkOrderedCompositeDistributor , 0.000119 seconds<BR> <BR>Server, Process 3<BR>Execute vtkFileSeriesReader id: 239, 0.000459 seconds<BR>Execute vtkPVGeometryFilter id: 306, 0.004905 seconds<BR>Execute vtkPVCacheKeeper id: 517, 6.9e-05 seconds<BR>Execute vtkMPIMoveData id: 520, 0.764289 seconds<BR>Dataserver gathering to 0, 0.740679 seconds<BR>Execute vtkOrderedCompositeDistributor , 0.000136 seconds<BR></SPAN> <BR>--------------------------------------------------------------------<BR> <BR> <BR>Unlike Case 2, Case 1 does not show any 'Dataserver gathering to 0' or 'Dataserver sending to client'<BR>so i guess what makes Case 2 slower is because the vtkMPIMoveData is doing extra thing to move data amongst processes<BR> <BR>btw, whenever i run it without -offscreen-rendering, i get this message:<BR>"Display is not accessible on the server side. Remote rendering will be disabled."<BR> <BR>based on what i read from paraview tutorial, what i understand is that the <FONT face=CMR12>parallel rendering algorithm allows each process to independently render its partition of the geometry and then </FONT><FONT face=CMBX12>composites </FONT><FONT face=CMR12>the partial images together to form the final image. Is this what happened in Case 2 because it has 'Dataserver gathering to 0' or 'Dataserver sending to client'?</FONT><BR> <BR>actually what i dont understand is why case 2 is much slower than case 1? <BR>is it because different 'things' are going on in the processes when offscreen-rendering is on/off?<BR> <BR> <BR>any help / reply is highly appreciated<BR> <BR>Best Regards,<BR>chewping<BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR></SPAN></BLOCKQUOTE><BR><BR><SPAN title=neteasefooter>
<HR>
<A href="http://allyes.nie.163.com/main/adfclick?db=afanie&bid=1260,614,23&cid=148,4,1&sid=1357&show=ignore&url=http://tx2.163.com/fab.html">看陆川杨幂新片《琴棋书画》，品网易3D国韵网游《天下贰》</A> </SPAN> </body>
</html>