Protected data spaces and container technologies facilitate access to AI and allow fair, secure and traceable data exchange.

Big Data Infrastructures

We develop technologies for the setting up of protected data spaces. It is not just the business community but society as a whole who can benefit from easier access to Artificial Intelligence (AI) applications, data, resources and infrastructures. Data spaces facilitate a fair, secure and transparent exchange of data.

We are developing technology cores which can be used to package Machine Learning tools into software containers which make using infrastructures such as these as easy and as flexible as possible. The containers can be operated in a variety of computing environments with very little effort and are easily scaled to different application scenarios.

Containerized software means we are also able to use hybrid Machine Learning processes and can link data-based and knowledge-based methods in one joint application package. Our architectures also support the automated deployment, scaling and administration of containers.

We have successfully used this technology in research and development projects, in laboratories for training courses and within experimental environments for our customers and partners. We have set up cooperative networks such as the “Fraunhofer Big Data and Artificial Intelligence Alliance”, the “International Data Spaces Association” and the “Big Data Value Association (BDVA)”.

Go to our publications

Research priorities

Platforms for scalable Machine Learning

We have developed a technology core and a dedicated laboratory for the research of Machine Learning algorithms and applications that can be scaled up for use with different applications.

Frameworks for scalable knowledge technology     

We expand and optimize existing Big Data frameworks in order that we can integrate semantic technology, particularly comprehensive knowledge graphs and information models.  

Data space technology

We are working in collaboration with the “International Data Spaces Association” on reference architecture for secure data ecosystems, their implementation and standardization.

Highlights

Reference architecture for secure data spaces

We are cofounders of the Germany-wide “Industrial Data Space” initiative, now operating worldwide as the “International Data Spaces Association”. Our aim is to make it possible for large multinational corporations as well as small and medium-sized companies to exchange data securely and easily throughout their organization. Our scientists have not only played a vital part in developing the reference architecture and information model for data rooms but also helped turn them into technologies and components.

Fraunhofer Big Data and Artificial Intelligence Alliance

The Fraunhofer Big Data and Artificial Intelligence Alliance is made up of 30 separate Fraunhofer institutes harnessing their expertise across all sectors. We not only help companies develop Big Data strategies, software packages and data protection compliant Artificial Intelligence systems but also train specialists and managerial staff to become “data scientists”.

Position paper

»Ökosysteme für Daten und KI«

The position paper outlines the data and AI opportunities offered up by ecosystems and recommends the action that needs to be taken. Fraunhofer projects in many different areas show the necessity for and the added value offered by data ecosystems. The paper was written on behalf of the Fraunhofer-Gesellschaft’s priority strategic initiative “Cognitive Systems, Artificial Intelligence and Data Sovereignty”. With Fraunhofer IAIS contributions. Publisher: Fraunhofer ISST and IAO. Language: German.

Urban data spaces study

Using Bonn, Dortmund, Emden and Cologne as examples the “Urbane Datenräume – Möglichkeiten von Datenaustausch und Zusammenarbeit im urbanen Raum” study illustrates how data management is currently being applied in communities. The study recommends an individually designed data room that can be easily and inexpensively added to a jointly used open platform core to improve the availability of urban data. The study was sponsored by the BMBF (German Federal Ministry for Education and Research) and carried out by scientists working for the Fraunhofer institutes FOKUS, IAIS and IML. Language: German.