The NIRD Central Data Library (CDL) is a new complimentary service for users of Sigma2’s national research infrastructure, in particular NIRD and the national HPC systems.
It is built on top of the NIRD Data Lake and acts as a central repository for data that needs to be reused and shared across projects.
CDL is suited for storing and sharing non‑persistent data such as:
- Input datasets enriched with metadata
- Shared data libraries
- AI‑models and AI‑training data
- Other reusable input data needed by multiple projects or collaborators
You can define access levels and user roles (e.g. data owners and depositors). Data is accessible via:
- S3 (default)
- POSIX and NFS from NIRD and Sigma2’s HPC systems
- Connections to third‑party systems and desktops
Data integrity is supported by daily snapshots, redundancy and error‑correcting mechanisms. CDL is not a long‑term archive and does not replace the NIRD Research Data Archive, and it currently does not include backup.
How to get access
You can request CDL resources at any time of the year by selecting the Central Data Library option in the application for NIRD storage resources. For details about the eligibility, and governance questions, please visit the Central Data Library policy.
Applications are evaluated based on technical requirements and scientific value, and approved projects are granted access accordingly.
More information about how to apply for resources on the national e-infrastructure.