Info page Large Scale Data Facility (LSDF3)
Storage Infrastructure for Research-Related IT Services.
The Large Scale Data Facility (LSDF) is a central storage infrastructure for large volumes of scientific data, operated by the University Computing Center and KIT. It serves as a reliable, professionally managed storage solution for numerous research-related services used by data-intensive disciplines—such as the life sciences, astrophysics, hydromechanics, medicine, scientific computing, the humanities, and the geosciences. Only through the use of a centralized storage infrastructure can the requirements for data volume, access speed, operational reliability, and sustainability be met.
Services for Researchers
Researchers can use specialized IT services at the Heidelberg campus that are based on the LSDF’s storage infrastructure and listed in the service catalog:
- SDS@hd – Scientific Data Storage: A state-level service for researchers in Baden-Württemberg for storing “hot” research data that is accessed frequently.
- heiARCHIVE: Central archiving service for Heidelberg University’s research data.
- Use of Hierarchical Storage Management (HSM): HSM-based Storage backend for repositories (in preparation): Storage backend for data publication services of various communities within the NFDI and in Baden-Württemberg.
Facts at a Glance
- Data Security: All data is copied to a disaster recovery system with media rotation. Defective media are destroyed on-site. Additionally, a snapshot mechanism is used (in the event of to protect against accidental deletion by users).
- Energy efficiency and cost efficiency through the use of hierarchical storage management (HSM): a combination of fast online storage and tape storage for large storage capacities.
- Capacity: The LSDF in Heidelberg has a usable capacity of over 20 PB of online storage and over 20 PB of tape storage.
- Integration with supercomputing resources: The LSDF is closely integrated with the bwForCluster Helix high-performance computing cluster, enabling the storage and processing of large volumes of data. This can also be accessed easily via the state-level service bwVisu.
- Strategic Research Infrastructure: The LSDF is embedded in the framework concept of the state universities for high-performance computing (HPC) and data-intensive computing (DIC) and is thus geared toward long-term viability and sustainability. Its strategic design and implementation are carried out in collaboration with KIT.
- Baden-Württemberg Data Federation: As part of the data federation in Baden-Württemberg, the LSDF provides a WebDAV endpoint and FTS3 access, thereby enabling high-performance transfers of large research datasets across locations.
- Funding: Infrastructure costs are financed through grants from the Ministry of Science, Research, and the Arts (MWK) and the German Research Foundation (DFG) as large-scale research facilities under Article 91b of the German Constitution (FUGG) and large-scale facilities of the federal states (LAGG), which are secured through a series of grant applications, most recently through LSDF3 and bwData4NFDI.