Use Paraview to interatively visualise and analyse large datasets using our computing resources.

Paraview runs in a client-server mode: the data is held and processed by a paraview server running inside a batch job which connects to a graphical client running on your personal computer. 

Configuring the Paraview Client

First download and install the Paraview client for WindowsmacOS or Linux.

Once Paraview is running,  choose File-Connect... or hit the connect icon directly. In the window that opens select "Add Server". Fill in the options:

Name: Imperial RCS
Server Type: Client/Server (reverse connection)
Port: 11111

and press OK.

Next set the Startup Type  to  Manual  and press Save, followed by Connect. Your client should open a pop-up window with a message saying Waiting for server to connect.In subsequent use simply select Imperial RCS from the Connect menu.

Starting the Paraview Server

The Paraview server must be run as a batch job. Before you can submit the job you will need to configure it with the IP address of the computer where you are running the client. If you don't know the IP address, you can look it up online.

If this is the first time that you are running Paraview, ensure that any firewall running on your computer has TCP port 11111 open.

Note that when you go to "Open ..." a file from your client you will see the filesystem as the server sees it.   Either have your data on shared filesystems (/work, etc) or have your job script copy the data to $TMPDIR.

Finally, submit a job using the following template:

            #PBS -lselect=1:ncpus=16:mem=62gb
            #PBS -lwalltime=4:0:0
            module load paraview/5.4.1
            mpiexec pvserver --reverse-connection --client-host=IPADDRESS 

replacing IPADDRESS with the IP you just determined, and optionally tailoring ncpus and mem to match your needs.

As soon as the job starts the Paraview Server will automatically connect to your Client.

If everything goes OK you should get the message "Client connected." and your client is now ready for your work.

If this fails, double-check your firewall settings, and - if you are outside the college network - retry after having connected to the VPN service.