The trend in large-scale research data management constitutes a paradigm shift from previous local data storage in file systems and databases to the coordinated orchestration of unstructured, high-volume data maintained in distributed sites.

Methods for data curation and analysis need to be capable of dealing with the three "v"s: volume, variety, and velocity of data. ZIB’s unique (in Berlin) expertise in development of mathematical methods, scalable algorithms, and flexible software for distributed data management, data curation, data analysis, and visualization will be combined to meet this challenge. Already today, ZIB’s research results and software prototypes are utilized in various scientific domains like medicine, biology, materials science, and earth system science. Domain-specific methods will be developed that allow for efficient analysis of very large sets of structured and unstructured data.