Panta rhei. This idea, which can be translated from Ancient Greek as "Everything flows," is attributed to the philosopher Heraclitus. What do you associate with it? If we were to ask our post-doc Sören Henning about it, the answer might be different than expected: Data streams.

At the beginning of May, Sören presented his paper "ShuffleBench: A Benchmark for Large-Scale Data Shuffling Operations with Distributed Stream Processing Frameworks" at the International Conference on Performance Engineering (ICPE), opens an external URL in a new window in London. Together with co-author Adriano Vogel and three other colleagues, he worked on this paper. The intensive work paid off: the paper was awarded as the best industry paper. "I am very happy that the need for research in this field has been recognized by the performance engineering community and that our work has been so positively received," says Sören.

With ShuffleBench, they introduced a new benchmark for evaluating distributed stream processing frameworks. These frameworks are essential for efficiently processing large amounts of continuous data streams. Unlike existing benchmarks, ShuffleBench focuses on scenarios where these frameworks redistribute data records for analysis, rather than the analysis itself. Adriano adds: "We are currently using and expanding ShuffleBench to provide the community with further quantitative insights, especially with regard to fault tolerance and performance optimizations in stream processing."
Want to learn more? Read the paper here: https://doi.org/10.1145/3629526.3645036, opens an external URL in a new window.

Sören and Adriano are both part of a jointly funded research project, opens an external URL in a new window between Johannes Kepler University and Dynatrace, opens an external URL in a new window Austria. This collaboration aims to explore current topics in computer science, particularly in the areas of software engineering, artificial intelligence, and data science. "The goal is to build a bridge between industry and academia and to advance innovative research together," explains Rick Rabiser. The success of this collaboration is evident. "I am very proud of the achievements of our JKU/Dynatrace Co-Innovation Lab," concludes Rick, happy about another award for his team.