09.04.2023: The hardware on Robin has been upgraded, replacing the
NVIDIA A30 GPUs with more powerful NVIDIA A40 GPUs. Up to 2 sessions
can now share a single GPU. See this link for additional information.
06.11.2023: the RStudio session is now available on the Robin cluster.
Check the description of RStudio sessions here for additional information.
16.10.2023: The Robin cluster and the Generic Jupyter Lab option have
been discussed during a Meet MPCDF presentation. The slides of the talk
are available here.
12.10.2023: A new Generic Jupyter Lab software has been added to
the options for the "Lab" interface. This re-designed Jupyter Lab interface
allows for several kernels to be available via Jupyter, without depending
on any Anaconda module available on the cluster.
Also, the default values in the "submit" page have been reorganized in
order to facilitate submissions on the new Robin cluster. In particular, default
values for the Jupyter interface and software have been modified: users are
encouraged to verify the selected parameters before requesting a session.
14.09.2023: The latest anaconda module (2023.03) is now available when
submitting a Jupyter session.
30.08.2023: The new Robin cluster is now available for users with access
to the Raven cluster (see this link for additional information).
Users interested in using the Remote Visualization Service on Robin are
reminded to initialize their sessions on the cluster once, before submitting
their first session.
16.05.2023: we are experiencing some problems with our 2FA servers. If
get the "cannot connect to any privacyIDEA-server" error during
login, please wait for the next OTP and use this newly generated token
to access the service.
27.10.2022: the "Advanced options" expandable section has
been added to the submission page. This section contains
costumizations for the selected session, if any is available.
Currently, only the possibility to prepend the Python path of
loaded modules in a Jupyter notebook is available.
Appending the Python path of modules loaded in the Jupyter
notebook remains the default.
