Sprecher
Beschreibung
At the German Aerospace Center in Augsburg, the data management system shepard (Storage for Heterogeneous Production and Research Data) is being developed as a tool for research data management. It organizes data and brings metadata to the data by organizational elements, starting from Collections as top level elements above Data Objects which in turn contain Data References. The latter point to the actual data and measurements, stored in dedicated containers. Shepard supports timeseries data, custom-tailored structured data, raw files and spatial data as data types. The use of semantic annotations allows for further enrichment with metadata.
Technologically, the core of shepard consists of an API, provided by a Quarkus server. A web based user interface allows for quickly viewing and browsing the content. However, in order to collect data automatically and general purpose scripting, the API has to be used. As a consequence, the hurdles for its usage should be as low as possible so that researchers do not have to spend too much time and effort to tap into the full potential of shepard.
In this session, we aim to showcase the low-threshold access to shepard. In particular, we present a one shot configuration which allows the user to build and run a local instance in short time. A script generating an example collection describing the inner solar system gives an introduction how to structure and to work with data in shepard. Furthermore, we show how the API can be used via Python clients also by rather inexperienced programmers. Particular focus is being given towards the integration of Jupyter notebooks as an easy possibility for visualization and first steps toward data evaluation. Also the frontend will be demonstrated where all kinds of objects can be created, related to each other, and semantically annotated in an easy intuitive way.
| Alternative Track | 2. Empowering Research Communities: Turning Metadata into Action |
|---|