Desktop Cloud Visualization (DCV)¶
NICE DCV enables remote accessing OpenGL 3D applications running on ZIH systems using the server's GPUs. If you don't need OpenGL acceleration, you might also want to try our WebVNC solution.
See the official DCV documentation if you want to know whether your browser is supported by DCV.
Access with JupyterHub¶
Check out our new documentation about Virtual Desktops.
To start a JupyterHub session on the partition
taurusi210[4-8]) with one GPU, six CPU cores
and 2583 MB memory per core, click on:
Optionally, you can modify many different Slurm parameters. For this
follow the general JupyterHub documentation.
Your browser should load JupyterLab, which looks like this:
Click on the
DCV button. A new tab with the DCV client will be opened.
Notes on GPU Support¶
- Check GPU support via:
marie@compute$ glxinfo name of display: :1 display: :1 screen: 0 direct rendering: Yes [...]
If direct rendering is not set to
Yes, please contact HPC support.
- Expand LD_LIBRARY_PATH:
marie@compute$ export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib64/nvidia