ALERT! Warning: your browser isn't supported. Please install a modern one, like Firefox, Opera, Safari, Chrome or the latest Internet Explorer. Thank you!
Startseite » ... » Zentrale Einrichtungen  » ZIH  » Wiki
phone prefix: +49 351 463.....

HPC Support

Operation Status

Ulf Markwardt: 33640
Claudia Schmidt: 39833

hpcsupport@zih.tu-dresden.de

Login and project application

Phone: 40000
Fax: 42328
servicedesk@tu-dresden.de

Desktop Cloud Visualization (DCV)

NICE DCV enables remote accessing OpenGL-3D-applications running on the server (taurus) using the server's GPUs. If you don't need GL acceleration, you might also want to try our WebVNC solution.

Note that with the 2017 version (and later), while there is still a separate client available, it is not necessary anymore. You can also use your (WebGL-capable) browser to connect to the DCV server. A standalone-client, which might still be a bit more performant, can be downloaded from https://www.nice-software.com/download/nice-dcv-2017

How to use:
  1. Log in to taurus.hrsk.tu-dresden.de via ssh
  2. Start interactive batch job on node taurusi2107 or taurusi2108 (which are both in partition gpu2-interactive) :
    > srun --pty --time=1:00:00 --gres=gpu:1 -w taurusi210[7-8] bash -i
    ... <wait for job allocation>
  3. Create a DCV session in the interactive shell:
    > dcv create-session MySessionName
  4. (optional) Confirm your session has been successfully created:
    > dcv list-sessions
  5. determine IP address of the node you have been assigned by the batchsystem (capital i):
    > hostname -I
  6. From your local desktop PC, start your web-browser and connect to the IP address from the previous step starting with 172 and port 8443, e.g. https://172.24.145.122:8443 and accept the SSL certificate
  7. Log in using ZIH credentials
  8. Try dcv: open a terminal (right-click, "Console") and call "dcvgltest" to see if it runs
  9. Start your OpenGL application and get your work done

Notes:
  • The visualization nodes have private IP addresses that are only routed within the campus network. If you are coming from an outside network, you have to either open a tunnel via ssh or use VPN, see https://tu-dresden.de/zih/dienste/service-katalog/arbeitsumgebung/zugang_datennetz/vpn
  • You can close your sessions explicitly via "dcv close-session MySessionName". When all your jobs on the node are finished, SLURM will clean-up for you anyways, though.
  • You can create multiple sessions, but at least the browser-based client does not allow you to select one, it will always connect to the one started last

Notes on GPU Support:
  • Check GPU support via:
glxinfo
name of display: :1
display: :1 screen: 0
direct rendering: Yes      # <--- This line!
...

If direct rendering is not set to yes, please contact hpcsupport@zih.tu-dresden.de
  • Expand LD_LIBRARY_PATH:
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib64/nvidia/