<P>
Thanks a lot for helping me with my frustrating condition. The problem<BR>
with the invisible window being created whose title in the task bar is<BR>
"unable to open display" has gone, I compiled the sources again with<BR>
off-screen rendering enabled and start the paraview server with<BR>
off-screen rendering now.<BR>
<BR>
Server end :<BR>
raashid@ender:~/src/cvs/paraview-unix-new/bin$ mpirun -np 2 ./pvserver --use-offscreen-rendering<BR>
<BR>
Client End :<BR>
raashid@ged:~/src/cvs/paraview-unix-new/bin$ ./pvclient --server-host=10.101.11.72<BR>
<BR>
The client gets connected to the server and works fine.<BR>
<BR>
You suggested to check the check-box next to Composite and move the<BR>
slider all the way to the left (0 MBytes) and then try to visualize.<BR>
When I do this no matter how small the dataset is when I open it and<BR>
press the accept button a message is displayed saying "Server<BR>
connection has been dropped. ParaView client will be terminated"<BR>
<BR>
The terminal output on client was :<BR>
<BR>
raashid@ged:~/src/cvs/paraview-unix-new/bin$ ./pvclient --server-host=10.101.11.72<BR>
Connected to server<BR>
ErrorMessage<BR>
# Error or warning: There was a VTK Error in file: /home/raashid/src/cvs/ParaView/Servers/Common/vtkServerConnection.cxx (67)<BR>
vtkServerConnection (0xc0ee1b8): Server Connection Closed!<BR>
ErrorMessage end<BR>
<BR>
The terminal output on server was :<BR>
<BR>
raashid@ender:~/src/cvs/paraview-unix-new/bin$ mpirun -np 2 ./pvserver --use-offscreen-rendering<BR>
Listen on port: 11111<BR>
Waiting for client...<BR>
Client connected.<BR>
-----------------------------------------------------------------------------<BR>
One of the processes started by mpirun has exited with a nonzero exit<BR>
code. This typically indicates that the process finished in error.<BR>
If your process did not finish in error, be sure to include a "return<BR>
0" or "exit(0)" in your C code before exiting the application.<BR>
<BR>
PID 6385 failed on node n0 (10.101.11.72) due to signal 11.<BR>
-----------------------------------------------------------------------------<BR>
<BR>
Also I tried 3 different cases of composting for a dataset of 0.5<BR>
million (vtk-xml file type, try to draw glyph at each point)<BR>
<BR>
Case 1. Composting is off <BR>
<BR>
Case 2. Composting is on and value is increased (to 10.0 MB)<BR>
<BR>
Case 3. Composting in on and value is turned to 0.0 MB<BR>
<BR>
In all cases I noticed one thing, on the lower left side of the<BR>
window, where the the running process is display, every time when the<BR>
message "MPIMoveData" is displayed the paraview crashes.<BR>
<BR>
I dont know if this thing is useful, but I just wanted to let you know. <BR>
<BR>
<BR>
On Mon, 09 Jul 2007 Moreland,Kenneth wrote :<BR>
>Raashid,<BR>
><BR>
><BR>
><BR>
>As I stated in an email I sent out last week (attached), it looks like<BR>
>ParaView may have the parallel rendering turned off, in which case all<BR>
>the geometry will be pushed to the client (which in turn would cause the<BR>
>slow rendering and memory issues).<BR>
><BR>
><BR>
><BR>
>Verify that parallel rendering is turned on. Go to 3D View Properties.<BR>
>Make sure that the checkbox next to Composite is checked and move the<BR>
>slider all the way to the left (0 MBytes). Also make sure that<BR>
>Subsample Rate is checked and move the slider to the right some. Now<BR>
>when you do an interactive render (render while you are moving the<BR>
>mouse), the image should look blocky. If it does not, parallel<BR>
>rendering is still off. Check the DISPLAY parameters on the server to<BR>
>ensure that it can open rendering windows.<BR>
><BR>
><BR>
><BR>
>-Ken<BR>
><BR>
><BR>
><BR>
>________________________________<BR>
><BR>
> From: paraview-bounces+kmorel=sandia.gov@paraview.org<BR>
>[mailto:paraview-bounces+kmorel=sandia.gov@paraview.org] On Behalf Of<BR>
>Raashid Baig<BR>
>Sent: Sunday, July 08, 2007 1:38 PM<BR>
>To: paraview@paraview.org<BR>
>Subject: [Paraview] Newbie question about paraview's parallel<BR>
>visualization procedure<BR>
><BR>
><BR>
><BR>
>I am a newbie to parallel visualization and ParaView in general and I<BR>
>am some very basic problems with ParaView.<BR>
><BR>
>I am using ParaView2.6.X and trying to do performance analysis with<BR>
>increasing dataset on ParaView as a part of an academic project. I<BR>
>want to see what are the gains in CPU & memory usage and time required<BR>
>to render a particular dataset when ParaView is runs in parallel. All<BR>
>machines have Intel Core Duo 2, 2GB DDR2 RAM, and 2GB swap, NVidia<BR>
>7series graphics cards and work on Debian/Ubuntu Linux.<BR>
><BR>
>I have structured data-sets with point attributes in form of VTK-XML<BR>
>files (*.vts). When I run paraview in stand-alone mode on a single<BR>
>machine, I am able to draw glyph at each point of the data sets upto 1.5<BR>
>million grid points. But when I put the same VTK XML files on 3<BR>
>different computers (all have ParaView 2.6.X installed) and run ParaView<BR>
>through LAM/MPI in client-server mode where server is running on 2<BR>
>computers and the client is running on another computer, performance<BR>
>decreases drastically. Not only does the time taken to render final<BR>
>image, cpu and memory usage increase, I am only able to dram glyps till<BR>
>0.5 million grid points. After 0.5 million point the get an error (this<BR>
>particular error is for a dataset of 0.5 million) :<BR>
><BR>
>on client :<BR>
><BR>
>------------------------------------------------------------------------<BR>
>---------<BR>
>raashid@ged:~/src/cvs/paraview-unix-new/bin$ ./pvclient<BR>
>--server-host=10.101.11.72<BR>
>Connected to server<BR>
>ErrorMessage<BR>
># Error or warning: There was a VTK Error in file:<BR>
>/home/raashid/src/cvs/ParaView/Servers/Common/vtkProcessModule.cxx<BR>
>(1171)<BR>
>vtkProcessModule (0xc0c7530): Received exception from server:<BR>
>Insufficient memory exception.<BR>
>ErrorMessage end<BR>
>ErrorMessage<BR>
># Error or warning: There was a VTK Error in file:<BR>
>/home/raashid/src/cvs/ParaView/Servers/Common/vtkServerConnection.cxx<BR>
>(67)<BR>
>vtkServerConnection (0xc0ee1b8): Server Connection Closed!<BR>
>ErrorMessage end<BR>
>------------------------------------------------------------------------<BR>
>---------<BR>
><BR>
>on master node of the server I get the following error message :<BR>
><BR>
>raashid@ender:~/src/cvs/paraview-unix-new/bin$ mpirun -np 2 ./pvserver<BR>
>--use-offscreen-rendering<BR>
>Listen on port: 11111<BR>
>Waiting for client...<BR>
>Client connected.<BR>
>Process id: 0 >> ERROR: In<BR>
>/home/raashid/src/cvs/ParaView/VTK/Common/vtkDataArrayTemplate.txx, line<BR>
>95<BR>
>vtkFloatArray (0xad19b48): Unable to allocate 46593000 elements of size<BR>
>4 bytes.<BR>
><BR>
>Process id: 0 >> ERROR: In<BR>
>/home/raashid/src/cvs/ParaView/Servers/Common/vtkProcessModule.cxx, line<BR>
>1187<BR>
>vtkProcessModule (0xa7f14c8): Exception: Insufficient memory exception.<BR>
><BR>
>Am I using the wrong file type, if yes what should I use ?<BR>
>I have not changed any LOD parameters. Is there some threshold limit I<BR>
>am violating ?<BR>
><BR>
><BR>
><BR>
>search<BR>
><http://adworks.rediff.com/cgi-bin/AdWorks/click.cgi/www.rediff.com/sign<BR>
>ature-home.htm/1050715198@Middle5/1236533_1230311/1276266/1?PARTNER=3&OA<BR>
>S_QUERY=null%20target=new%20><BR>
><BR>
><BR>
><BR>
>Raashid,<BR>
><BR>
><BR>
><BR>
>First, let me assure you that ParaView's parallel visualization and<BR>
>rendering are efficient and scalable. We have been able to use ParaView<BR>
>to visualize 6 billion cell grids and have clocked rendering speeds of<BR>
>over 8 billion polygons per second (albeit with much larger clusters<BR>
>than you have set up).<BR>
><BR>
><BR>
><BR>
>As far as your problems are concerned, I'm going to have to do a lot of<BR>
>guessing because I cannot replicate your problem and there are missing<BR>
>pieces of information. Looking at your mpirun command, it appears that<BR>
>the DISPLAY variable is set to the same value on every process. That<BR>
>is, all processes in the ParaView server are creating a remote (for<BR>
>them) X window on 10.101.11.71. If that is the case, it is very, very<BR>
>bad. Basically, that would mean that all of the rendering is being<BR>
>funneled into one computer. Thus, instead of actually getting parallel<BR>
>rendering, you are getting serial rendering plus the added overhead of<BR>
>moving OpenGL primitives and images across the network plus the standard<BR>
>overhead of parallel rendering. Note that because you are only using 3<BR>
>computers, I would not expect the parallel gains to be much higher than<BR>
>the natural overhead of the parallel algorithms. If your computers are<BR>
>using a standard Ethernet connection, the relatively high latency and<BR>
>low bandwidth will further erode the parallel gains.<BR>
><BR>
><BR>
><BR>
>The fact that your parallel job is crashing when you have a parallel job<BR>
>with a sizeable data set suggests that you might not be using parallel<BR>
>rendering at all. You can shut off parallel rendering with the 3D view<BR>
>properties or ParaView will shut it off for you if the server cannot<BR>
>create the appropriate rendering windows. When parallel rendering is<BR>
>shut off, ParaView does the only thing left: funnel all of the geometry<BR>
>to the client.<BR>
><BR>
><BR>
><BR>
>Verify that you are actually doing parallel rendering. Start by running<BR>
>in client/server mode (don't bother with parallel standalone: its<BR>
>usefulness is basically limited to debugging). Go to 3D View<BR>
>Properties, verify that the checkbox next to "Composite" is checked and<BR>
>move the slider all the way to the left (0 MBytes). If everything is<BR>
>working right, parallel rendering should now always be used. The<BR>
>easiest way to verify that is to observe the subsampling, which only<BR>
>happens on interactive rendering when rendering in parallel. Make sure<BR>
>that the checkbox next to "Subsample Rate" is checked and move the<BR>
>slider to the right to 4 Pixels or more. Now when you rotate your 3D<BR>
>rendering, it should look pixilated.<BR>
><BR>
><BR>
><BR>
>Offhand, I cannot think of another reason for your memory failures. I<BR>
>just used the Wavelet source to create a 125,000,000 cell structured<BR>
>mesh on three nodes of our visualization cluster (each with 1 MB each),<BR>
>then created 5 isosurfaces (~7.4 million cells and ~3.7 points) and<BR>
>rendered it just fine.<BR>
><BR>
><BR>
><BR>
>One last piece of advice: don't run D3 on your structured data.<BR>
>ParaView can balance your structured data without any help. D3 is<BR>
>designed for unstructured data. The first thing it will do is convert<BR>
>the structured data to unstructured, which is a much less efficient<BR>
>representation. It makes more sense to run D3 on the output of the<BR>
>contour filter, but that is probably not worthwhile unless the contour<BR>
>is very big and very unbalanced.<BR>
><BR>
><BR>
><BR>
>-Ken<BR>
><BR>
><BR>
><BR>
> _____<BR>
><BR>
> From: paraview-bounces+kmorel=sandia.gov@paraview.org<BR>
>[mailto:paraview-bounces+kmorel=sandia.gov@paraview.org] On Behalf Of<BR>
>Raashid Baig<BR>
>Sent: Thursday, July 05, 2007 12:25 PM<BR>
>To: paraview@paraview.org<BR>
>Subject: [Paraview] Newbie question about paraview's parallel<BR>
>visualization procedure<BR>
><BR>
><BR>
><BR>
>I am a newbie to parallel visualization and ParaView in general, I<BR>
>have some doubts which may be very fundamental.<BR>
><BR>
>I am using ParaView2.6.X and trying to do some performance analysis<BR>
>and dataset limit of ParaView in general as a part of an academic<BR>
>project. I want to see what are the gains in CPU & memory usage and<BR>
>time required to render a particular dataset when ParaView is run in<BR>
>parallel. All machines are Intel Core Duo 2, 2GB DDR2 RAM, and 2GB<BR>
>swap, all machines have NVidia 7series graphics cards and have<BR>
>Debian/Ubuntu Linux<BR>
><BR>
>I have structured data-sets with point attributes in form of VTK<BR>
>legacy files. When I run paraview in stand-alone mode on a single<BR>
>machine, I am able to draw 5 iso-surface for structured grid datasets<BR>
>upto 5 million points. But when I put the same VTK legacy files on 3<BR>
>different computers (all have ParaView 2.6.X installed) and run<BR>
>ParaView through LAM/MPI in distributed stand alone mode (as described<BR>
>in the ParaView book) on these 3 machines, the performance decreases<BR>
>for the isosurface case. Not only does the time taken to render final<BR>
>image, cpu/memory usage increase, I am only able to visualize till 1<BR>
>million grid points. After 1 million point the get an error (this<BR>
>particular error is for a dataset of 2 million) :<BR>
><BR>
>raashid@ged:~/src/cvs/paraview-unix/bin$ mpirun -np 3 -x<BR>
>DISPLAY=10.101.11.71:0.0 ./paraview<BR>
>ErrorMessage<BR>
># Error or warning: There was a VTK Error in file:<BR>
>/home/raashid/src/cvs/ParaView/VTK/Common/vtkDataArrayTemplate.txx (266)<BR>
>vtkIdTypeArray (0xe041768): Unable to allocate 59135376 elements of size<BR>
>4 bytes.<BR>
>ErrorMessage end<BR>
>------------------------------------------------------------------------<BR>
>-----<BR>
>One of the processes started by mpirun has exited with a nonzero exit<BR>
>code. This typically indicates that the process finished in error.<BR>
>If your process did not finish in error, be sure to include a "return<BR>
>0" or "exit(0)" in your C code before exiting the application.<BR>
><BR>
>PID 16283 failed on node n0 (10.101.11.71) due to signal 11.<BR>
>------------------------------------------------------------------------<BR>
>-----<BR>
><BR>
>I get the same error as above if I use a D3 filter before applying the<BR>
>iso-contour.<BR>
><BR>
><BR>
>Following the ideas of a technical report I tried to open the VTK<BR>
>legacy file when ParaView server was running on 3 different computers<BR>
>and the client was running on a different computer, apply the D3<BR>
>filter and then save the file as PVTK unstructured grid format<BR>
>(*.pvtu, with 1 ghost levels), close the paraview, start a fresh<BR>
>paraview session with the same configuration (ParaView server was<BR>
>running on 3 different computers and the client was running on a<BR>
>different computer) load the master *.pvtu file and do the same<BR>
>iso-contour example as above, the server again crashes and displays<BR>
>the following error.(this particular error is for 2.5 million grid<BR>
>points)<BR>
><BR>
>On the server side I get the message :<BR>
>------------------------------------------------------------------------<BR>
>------<BR>
>raashid@ender:~/src/cvs/paraview-unix/bin$ mpirun -np 3 -x<BR>
>DISPLAY=10.101.11.71:0.0 ./pvserver<BR>
>Listen on port: 11111<BR>
>Waiting for client...<BR>
>Client connected.<BR>
>Process id: 0 >> ERROR: In<BR>
>/home/raashid/src/cvs/ParaView/VTK/Common/vtkDataArrayTemplate.txx, line<BR>
>95<BR>
>vtkIdTypeArray (0xd243258): Unable to allocate 75216436 elements of size<BR>
>4 bytes.<BR>
><BR>
>Process id: 0 >> ERROR: In<BR>
>/home/raashid/src/cvs/ParaView/Servers/Common/vtkProcessModule.cxx, line<BR>
>1187<BR>
>vtkProcessModule (0xb395560): Exception: Insufficient memory exception.<BR>
>------------------------------------------------------------------------<BR>
>------<BR>
><BR>
>On the client side I get the message<BR>
>------------------------------------------------------------------------<BR>
>-------<BR>
>raashid@ged:~/src/cvs/paraview-unix/bin$ ./pvclient<BR>
>--server-host=10.101.11.72<BR>
>Connected to server<BR>
>ErrorMessage<BR>
># Error or warning: There was a VTK Error in file:<BR>
>/home/raashid/src/cvs/ParaView/Servers/Common/vtkProcessModule.cxx<BR>
>(1171)<BR>
>vtkProcessModule (0xc0c7530): Received exception from server:<BR>
>Insufficient memory exception.<BR>
>ErrorMessage end<BR>
>ErrorMessage<BR>
># Error or warning: There was a VTK Error in file:<BR>
>/home/raashid/src/cvs/ParaView/Servers/Common/vtkServerConnection.cxx<BR>
>(67)<BR>
>vtkServerConnection (0xc0ee1b8): Server Connection Closed!<BR>
>ErrorMessage end<BR>
>------------------------------------------------------------------------<BR>
>--------<BR>
><BR>
>Am I using the wrong file type, if yes what should I use ?<BR>
>Is there some threshold limit I am violating ?<BR>
><BR>
>Please help me with this problem, because I am in serious trouble.<BR>
><BR>
><BR>
><BR>
</P>
<br><br>