JupyterHub changes

During the downtime of the cluster we also changed to a new JupyterHub version. Please see below the changes and new features.

The current system also does not run in a python virtual-environment anymore, which will allow own python modules. Please be aware that these still need to be compatible with the running version (3.9.9).

Run notebooks on the cluster

The system again supports running notebooks/labs on cluster nodes, the local server should only be used for testing, also most of the (new) features will not work on the local server.

Currently this system is part of the new SLURM 20 installation, so only a few nodes (and only a single GPU node) are available, but this will change soon.

Default environment JupyterLab

We changed the default environment to JupyterLab, the “next generation” user interface for Jupyter.

You can certainly switch to the classic notebook by selecting “Help” and “Launch Classic Notebook”.

Other features

Integrated Slurm Queue Manager, which allows you to see/manage your jobs.

Collaborative sharing and working on notebooks.

If you would like to use this feature, you need to send an email to it@ist.ac.at, this feature is not automatically enabled.

R-Studio is available

It is now also possible to run R-Studio Server within your Jupyter-Session. Just click on of the Rstudio Icons to start you server. Please be aware, that this system is new, so we are happy to receive your feedback via it@ist.ac.at.