Big Data Architecture and Analytics

Big Data technologies are indispensable for data that is too big, too fast, and too complex to be processed by classical techniques. We investigate and set up high-performance infrastructures for complex data analytics tasks on very large datasets or under realtime conditions.

Going beyond the pure infrastructure, we also investigate new data mining algorithms - such as stream mining or sub-linear algorithms - which scale under these extreme new conditions.

Living Lab

As a platform for learning and experimentation with big data applications we set up a living lab. We chose the lambda architecture as a flexible, highly scalable framework. It is suited for the analysis of streaming data in real-time, as well as for voluminous, but less time-critical batch analyses. The lab uses map-reduce techniques, parallel workflows, message passing and NoSQL databases.