Connecting to JupyterHub, requesting resources (RAM, processors, GPUs) for your Jupyter session, and starting your session

Connecting

From a web browser navigate to the following URL from a machine that is on-campus or has a VPN connection to the University: 

https://mesoshared.univ-fcomte.fr

When prompted, log in using your mesoFC credentials (username/password).

Click Start server when prompted:

Requesting resources (Spawner options)

Before the Notebook server starts you may wish to request resources such as memory, multiple CPUs and GPUs using the Spawner options form. We provide 4 templates:

  1. Local host process: spawn local process, used for testing
  2. Request 1 CPU, 4Go memory for 2hours: run SGE Job in background
  3. Request 8 CPU, 32Go memory for 2hours: run SGE Job in background
  4. Request 1 CPU, 1GPU, 16 Go memory for 2 hours: run SGE Job in background

Starting Notebook session

After you have specified the resources you want for your job, click Spawn to try starting a Jupyter session on one (or more) worker nodes.

If the cluster is busy or you have requested an incompatible or otherwise unsatisfiable set of resources from the job scheduler then this attempt to start a session will time out and you will return to the Spawner options form.

Once your session has started you should see the Jupyter file browser interface: