|This page is scheduled for deletion because it is either redundant with information available on the CC wiki, or the software is no longer supported.|
- 1 Overview
- 2 Versions
- 3 Usage
- 4 Notes
- 5 Paraview on Graham and Cedar
- 6 References
|Note: Some of the information on this page is for our legacy systems only. The page is scheduled for an update to make it applicable to Graham.|
|Description: An open-source, multi-platform data analysis and visualization application.|
|SHARCNET Package information: see PARAVIEW software page in web portal|
|Full list of SHARCNET supported software|
Sharcnet provides paraview for remote graphical use over a web based connection from the sharcnet systems page as described in the graphical section below.
There is no paraview module. Instead the latest version of paraview installed will be found by running the paraview command.
Paraview is not installed on the clusters.
There is no command line usage component for paraview at this time.
After connecting to visualization workstation via the Visualization Systems table found on sharcnet system page (as described in https://www.sharcnet.ca/help/index.php/Remote_Graphical_Connections) one can start ParaView by clicking Applications Menu -> graphics -> Paraview Viewer.
ParaView can also be started by opening a terminal emulator and typing paraview enter.
Note that adding a --stereo option will enable stereo rendering when supported by the hardware.
Running Paraview In Client/Server Mode
While running Paraview remotely on viz station is the preferred way, you can also run it in client/server mode. To do so, you need to download and install ParaView for your local computer (Linux, MacOS, and Windows version exist) from the ParaView Website (or install it using your package manager under Linux). Note that you must
- use the same version of ParaView as on the visualization stations, and
- setup ssh for passwordless login, and
- setup Paraview server configuration on your local computer (see examples for different operating systems)
Then, you are ready to run Paraview in client/server mode
- launch Paraview on your local computer
- click the connect icon (third from the the right) or picking connect from the file menu will now bring up a menu that allows you to connect to SHARCNET. Pick SHARCNET and then clicking connect. This will give a menu where you can pick the number of CPUs (more can actually be slower) and enter your SHARCNET login.
If everything goes correctly, you should see csrc://localhost:<number> (where <number> is some number) in your pipeline browser. Data files will now be your SHARCNET files and only state files will be local. Beware that any changes made to servers.pvsc while ParaView is running will be lost.
Set Up Configure File For The Paraview Server
Linux and MacOS
Place the following in servers.pvsc under .config/ParaView in your home directory (this must be done while ParaView is not running)
<Servers> <Server name="SHARCNET" resource="csrc://localhost"> <CommandStartup> <Options> <Option name="PV_SERVER_HOST" label="Host" save="true"> <String default="vdi-fedora20.user.sharcnet.ca"/> </Option> <Option name="PV_SERVER_PORT" label="Port"> <Range type="int" min="1024" max="65535" step="1" default="random"/> </Option> <Option name="PV_NODES" label="CPUs" save="true"> <Range type="int" min="1" max="16" step="1" default="4"/> </Option> <Option name="PV_USERNAME" label="Login" save="true"> <String/> </Option> <Option name="PV_CONNECT_ID" label="ID" readonly="true"> <Range type="int" min="1" max="65535" step="1" default="random"/> </Option> </Options> <Command exec="ssh" timeout="0" delay="5"> <Arguments> <Argument value="-x"/> <Argument value="-R"/> <Argument value="$PV_SERVER_PORT$:localhost:$PV_SERVER_PORT$"/> <Argument value="-l"/> <Argument value="$PV_USERNAME$"/> <Argument value="$PV_SERVER_HOST$"/> <Argument value="sh"/> <Argument value="-c"/> <Argument value="'module load mpi/openmpi-x86_64; DISPLAY=:0 exec mpirun -n $PV_NODES$ pvserver --reverse-connection --client-host=localhost --server-port=$PV_SERVER_PORT$ --connect-id=$PV_CONNECT_ID$'"/> </Arguments> </Command> </CommandStartup> </Server> </Servers>
A similar file, customized for your ssh setup, must be placed under your applications settings directory (C:\Users\YOUR-USER-NAME\Application Data\ParaView), and then the above directions will apply as well.
Server Connection Fails
You need to log on the viz station you want to connect to via VNC so ssh can trigger mounting your home directory. In order to fully bring up the graphics-related environment that is needed by Paraview, you need to use VNC before making connection from within Paraview.
Paraview Seg Faults On Startup
Paraview currently must be run under VirtualGL to work. This is configured by default under VNC. To do so over a ssh -Y requires running it under the vglrun -c proxy command.
[roberpj@gulper:~] ssh -Y firstname.lastname@example.org [roberpj@viz1-uwo:~] paraview Segmentation fault (core dumped) [roberpj@viz1-uwo:~] vglrun -c proxy paraview
While this paraview will run with this approach, it is not recommended. It will be very slow compared to connecting by VNC.
Paraview on Graham and Cedar
Please refer the visualization wiki page on CC for instructions about how to run Paraview and other visualization packages (VisIt, VMD) on the national systems.
o Paraview Help Documentation
o Sharcnet Paraview Training Document