...
The European Centre for Medium-Range Weather Forecasts (ECMWF) operates Data Store Services (DSS) as a family of common architecture data stores, such as the Climate Data Store (CDS), Atmosphere Data Store (ADS), Early Warning Data Store (EWDS), etc. To this effect, the DSS provides access, among others, to the data products produced and brokered as part of, the Copernicus Climate Change Service (C3S), the Copernicus Atmospheric Monitoring Service (CAMS), the Copernicus Emergency Management Service (CEMS) and internally produced ECMWF data products. In addition to the data access, the ECMWF provides a separate JupyterHub service for DSS (DSS JupyterHub Service or Service), by means of which DSS users can have interactive compute sessions (JupyterLab sessions) on DSS compute infrastructure, typically associated with with Python
| Footnote |
|---|
Copyright © 2001-2024 Python Software Foundation; All Rights Reserved, License terms available at https://docs.python.org/3/license.html. |
...
The use of JupyterHub in DSS is to enable the deployment of compute sessions (JupyterLab sessions) to DSS users based on availability. These JupyterLab sessions provide fast access to the data available on the various data stores and allow DSS users to perform post-processing and visualisation of this data. These JupyterLab sessions are not designed for very large computations
| Footnote |
|---|
For larger computation tasks, CDS users may consider other JupyterHub resources, for example WEkEO. |
. For clarification JupyterLab is a flexible and extendible user interface for interactive computing. It has several extensions that are tailored for using Jupyter Notebooks, as well as extensions for other parts of the data science stack.
The use of JupyterHub in DSS as described in these Terms of Use is hereinafter referred to as ‘DSS JupyterHub Service’ or ‘Service’.
...
Once logged in, DSS users have a choice of environment to use for their DSS JupyterHub Service JupyterLab session, as described below in Article (4) Description of the Service.
...
DSS users are able to spawn JupyterLab sessions, by means of the DSS JupyterHub Service, with one of the environments summarised in the environments available to DSS users section in the complete technical description of the DSS JupyterHub Service.
To ensure fair usage of DSS JupyterHub Service and/or the respective data store, ECMWF reserves the right to prioritise smaller (and shorter) JupyterLab sessions before others.
Pre-installed Software
...
. The packages included in the Pre-instatlled environments are open-source and described in the Pre-installed software section of the complete technical description of the DSS JupyterHub Service.
DSS users can install additional packages from the conda-forge channel (`conda install PACKAGE-NAME`), or from PyPi (`pip install PACKAGE-NAME`). These packages will be installed in the DSS user’s Private Storage and will be available next time the DSS user creates a JupyterLab session. For such packages and other software, additionally installed by the DSS user Articles 5.3-5.5 below shall apply.
...
If a Private Storage is not accessed (i.e., by spawning a JupyterLab session) for a period of thirty-one (31) consecutive days, this Private Storage together with its contents may be deleted in its entirety.
...
The purpose of the Scratch Storage is to store files for a current JupyterLab session or several JupyterLab sessions and such files shall not be considered permanently stored. The Scratch Storage is mounted on a shared disk and is cleaned regularly to ensure that the disk does not exceed capacity. Therefore, the lifetime of the storage of files in the Scratch Storage depends on the overall usage of the DSS JupyterHub Service by all DSS users. The cleanup of files is based on the least recently modified files.
...
ECMWF may at its sole discretion, under any circumstances, for any or no reason and with or without prior notice to the DSS users, suspend, terminate or restrict any DSS user’s access to the DSS JupyterHub Service, particularly in cases of violation of these Terms of Use or any of the applicable license terms or if a Private Storage is not accessed (i.e., by spawning a JupyterLab session) for a certain period of time beyond thirty-one (31) consecutive days.
...