IT Center Blog

Background knowledge on the research data storage

September 24th, 2021 | by
Cyperspace data cable electronic

Source: Pixabay

We already reported on the commissioning of the new research data storage system at the beginning of the year. In this article, we take a closer look at the process level of the storage system.

Technical representation of the storage system

The organization of the storage system into local storage pools (Aachen, Cologne, Bochum, Dortmund, Duisburg Essen) and the joint NRW pool (Aachen, Cologne and Duisburg Essen) is substantial for this.

Like the NRW pool, the local pools are each based on several systems, which are also located in different buildings. The legal ownership of the individual IT systems always lies with the university where they are installed. By distributing the IT systems of a pool over several buildings, increased data security is implemented.

The local pools are also managed by the respective university where the systems are installed. The design of the associated processes around storage usage is determined by local requirements and structures.

The NRW pool, on the other hand, is managed jointly and is used for the longer-term storage of data that requires special protection. The greater geographical distance between the installation sites means that the data is particularly well protected against local disasters.

Process level: Offerings for researchers in Aachen

In Aachen, the storage system will initially be usable via the FDM platform Coscine. This makes it possible to ensure that all data is described with metadata. In addition, the memory is always assigned to a project context.

Furthermore, Coscine offers the possibility of storing data securely over the long term. Thus, the platform fully supports the requirements of Good Scientific Practice as well as the standards of modern research data management. Furthermore, Coscine has programming interfaces (APIs) that allow the storage system to be addressed as part of automated applications and processes.

In addition, direct access to the storage was realized via the S3 protocol.

When using the S3 protocol directly, however, there is no support for structuring and annotating the data, so this is entirely the responsibility of

the users. For this reason, they must describe in a data management plan how the securing of the data is implemented in accordance with the standards of modern research data management. More information about the FDS resources and how to get them has already been described in another article.

Responsible for the content of this article is Jana Baur.

Comments are closed.