The research of this working group focuses on scalability, reliability and efficiency of large scale systems from an algorithmic and practical point of view. For example, we study the efficient analysis of large graphs in the context of SAT solving and heuristic search, as well as the design of systems without a single point of failure such as structured peer-to-peer overlays, and distributed and parallel algorithms in general.
We try to transfer our knowledge of the resilience and scalability of distributed systems to parallel applications to make large scale parallel applications efficient and more robust against (partial) system failures. The group focuses on all aspects of scalability, reliability and efficiency in parallel and distributed systems especially for next-generation processor, memory, and network technologies.