There is a dedicated visualization server called viz01. It mounts the Lustre file system, has 64 GB of RAM and 4 AMD Opteron 8132 processors, each with 4 processor cores, as well as two NVidia Quadro FX5600 graphics cards. For security reasons, this server does not have an IP address visible from outside the CHPC. However, it can be accessed via the login node by means of SSH tunneling. Although it is possible to perform remote visualization by means of X-forwarding (log in with “ssh -X”), this approach is generally too slow to be of use if the user is not on the internal network. It is therefore preferable to use a remote desktop. However, standard VNC is to slow for this and does not work properly with OpenGL, the library mostly used for 3D graphics. The visualization server has been set up with TurboVNC and VirtualGL, which gets around both of these problems. Use the following process for remote visualization:

• Log in to the system by normal means (command line ssh or PuTTy)
• From the cluster login node, log in to viz01, by means of ssh
• On viz01, start up a TurboVNC server for your use. You can either leave this login session open or disconnect after starting the TurboVNC server, which will continue running untill it is manually shut down.
• From your own system, set up an ssh tunnel to forward a port on your system to the appropriate port on viz01
• Use the TurboVNC client to connect to the VNC server on viz01
• Run your graphics program with vglrun
• When finished, close down the TurboVNC client, ssh into viz01 again and kill the TurboVNC server session

There is a default VNC server installed on viz01. Do not use it. Use TurboVNC, which can be started with the following command:

/opt/TurboVNC/bin/vncserver :3 -geometry 1600×900 -depth 24

More than one VNC session can be run simultaneously. In this example, two other sessions are already running, therefore :3 is used to specify that this will be a session on virtual display 3. If this is unavailable, try :4, etc. Optionally specify an appropriate display resolution and colour depth, as in the above example.