Gast ::
| Hauptseite > Publikationsdatenbank > Jupyter4NFDI - a central Jupyter Hub for the NFDI |
| Hauptseite > Publikationsdatenbank > Jupyter4NFDI - a central Jupyter Hub for the NFDI |
| Talk (non-conference) (Other) | FZJ-2024-06256 |
2024
This record in other databases:
Please use a persistent id in citations: doi:10.5281/ZENODO.14162496 doi:10.34734/FZJ-2024-06256
Abstract: Jupyter Notebooks are widespread across scientific disciplines today. However, their deployment across various NFDI consortia currently occurs through individual JupyterHubs, resulting in access barriers to computational and data resources. Whereas some of the services are widely available, others are barricaded within VPNs or otherwise inaccessible for a wider audience. Our ambition is to improve the user experience by offering a centralized service to extend the reach of Jupyter to a broader audience within the NFDI and beyond. This will be facilitated through two main constituents. Firstly, the technical foundation for our service will be the versatile configuration frontend that has been proven to meet user needs for the past seven years at JSC. It is continuously extended and traces and ever growing set of backend resources ranging from Cloud based, small-scale JupyterLabs to full-scale remote desktop environments on high-performance computing systems such as Germany's highest-ranked TOP500 system JUWELS Booster. We envision to extend this approach to various backend systems, such as the ones provided by the NHR. Secondly, an integration into the developing IAM4NFDI base service, facilitating user and community access to the above mentioned resources. Given the diversity of resources and their individual policy for user access, a central base service must support its users in the best possible way to access only those resources available to them. We plan to achieve this through appropriate user guidance, including information about how to apply for resources. Moreover, we will technically ensure that users can only see and configure those resources to which they have access.Importantly, the centralized system will not only simplify access but also support the import of projects along with their necessary dependencies, fostering an ecosystem conducive to creating reproducible FAIR Digital Objects (FDOs), possibly along with notebook identifiers supported by PID4NFDI. Our commitment is to a collaborative and centralized approach that brings Jupyter notebooks within everyone's reach and democratizes access to the infrastructures for reproducible computational research within the NFDI and beyond. In this talk, we'll revisit the history of the current solution, the landscape in which we intend to make it available, and give an outlook on the future of the service.
Keyword(s): Jupyter ; FAIR ; NFDI
|
The record appears in these collections: |
PDF