Vorschau

PDF, Englisch - Hauptdokument Download (10MB) | Nutzungsbedingungen

Abstract

Many Research Data Management (RDM) services are not confined to a single scientific domain or institution. Since no individual research institution can fully support all aspects of RDM with the required depth and domain specific details for every discipline, collaboration and shared services are both logical and necessary. This approach aligns with the broader objective of OneNFDI within the National Research Data Infrastructure, which aims to create a cross-disciplinary RDM landscape that meets the specific needs of research groups and disciplines, while also enabling the (re-)use of data across traditional boundaries as well as the deployment of AI approaches. In line with this vision, one of the long standing core objectives of the DataPLANT consortium is to provide tools and services that can be shared with other consortia and deployed across flexible backend infrastructures. At the heart of these technical services is the PLANTdataHUB, a science gateway designed to support data management and publication workflows. In addition to core functions, the PLANTdataHUB offers workflow pipeline templates, automated quality assurance, and other interaction services to foster DataPLANT's open participation and contribution model. This model encompasses all relevant assets, including ARCs, raw data, metadata templates, ontologies, code, and workflow descriptions. The backend services are designed as flexible, cloud-based microservices, supporting both on-premises installations and future integration with a shared NFDI infrastructure. The PLANTdataHUB thus provides a range of RDM workflows to support data scientists throughout different stages of the research data lifecycle, from development through to the publication of results.

The development of applications and tools for bottom-up, community-driven research data management requires the collaboration of multiple stakeholders. In building these services, we have adhered to design principles that provide high-level guidance and ensure the creation of sustainable and maintainable applications. In DataPLANT, tool development is driven by community needs, communicated through researchers (e.g., via data stewards) to developers. The process follows an incremental and iterative approach, ensuring continuous stakeholder alignment and commitment. Over the past four years, the DataPLANT team has developed and extended a suite of software and system components like GitLab or InvenioRDM tailored to enhance the digital infrastructure of plant scientists. These core services focus on data management, versioning, sharing, and publishing, and are designed as portable modules that can be integrated into a broader base infrastructure. A reverse proxy automatically manages SSL certificates for new services, significantly enhancing the flexibility of the infrastructure by enabling rapid changes to backend services without altering the static user-facing service hostnames.

Key design priorities include security, re-deployability, and fast recovery in case of system failures or infrastructure disruptions. The services integrate with a central authentication system based on the Life Sciences AAI and ORCID, with future compatibility planned for IAM4NFDI. In addition to the core components, monitoring services detect service outages, while logging services track performance data, access patterns, and accounting information. Through these efforts, we aim to encourage broader adoption by other communities and integration into the future NFDI service landscape.