Information infrastructure

Reproducibility of research and FAIR (Findable, Accessible, Interoperable, Reusable) data principles are essential parts of good scientific practice. Research data and information management is one prerequisite of this. In the first funding period, the service project INF built up an integrated data and information infrastructure for the whole TRR 280, and supported the researchers in its use. The infrastructure includes tools for the management and processing of numerical, simulation, mechanical, computer tomography, and other data. We placed particular emphasis on implementing best practices in the data life-cycle. This involves facilitating structured data storage for researchers, metadata annotation for accurate description and use, and ensuring that data is available and searchable.

In the second funding period, the aim of the INF project is to extend the capabilities of the current infrastructure. More emphasis will be put on the processes for long term archiving, automated data quality control, and data export to open science repositories. New models, data types, and workflows from the new research projects have to be integrated and implemented in the system.

The INF project will operate the whole central research data infrastructure, provide support and training on research data management and the utilized tools, as well as further develop and adapt the infrastructure to the needs of the researchers. Furthermore, the ongoing work in the German National Research Data Initiative (NFDI), especially in the consortium NFDI4Ing, is closely monitored, and suitable services developed by the NFDI consortia will be utilized or implemented as required. Another objective is to assist the research projects in utilizing AI and machine learning for data analysis and construction in collaboration with the national competence center for Big Data and Artificial Intelligence ScaDS.AI Dresden/Leipzig. Together with the researchers, the INF project will investigate the requirements on AI/ML, help in the implementation, provide performance support and access to the necessary HPC resources.