The provision of powerful computing performance is a key requisite for the implementation of complex systems and applications in science and technology. Today, high-performance computers generally take the shape of parallel computers. In such systems, a complex task is not processed by one single computer, but by several computers simultaneously. These share the workload in a similar way to an assembly line or a large company with different business areas. The parallel computing performance can be produced both by a single computer, comprised of several processors, and by several computers, that are distributed at differents sites and communicating with each other.
The analysis and design of efficient parallel and distributed computer architectures, the development of powerful methods of implementing applications on those systems, and implementing prototypes of those applications are essentially the areas covered by the "Parallel Computing" research area.
One major field of applications is that of computer simulation by parallel computers. Expensive and time-consuming test series, or dangerous experiments, can increasingly be replaced by computer simulations. The visualisation of 3-dimensional objects in real-time is of enormous relevance both to computer simulation and to a wide range of other applications. The necessary computing power can only be provided by means of scalable parallel computers. Such great computing power is also needed for solving decision-making problems in a wide variety of planning issues such as how valuable resources can be saved or systems used more efficiently in traffic regulation. The methods, we develop, are especially tested in the field of parallel chess programming. The Paderborn chess program P.ConNers was the first chess program in the world to win an official Grandmaster chess tournament (July 2000). Applications using parallel and distributed computing are also found in the field of networked multimedia systems, where memory and computing performance is made available "on demand".
Against this background; our main research areas are:
- The theoretical fundamentals of parallel computing;
- the architecture and operation of parallel and distributed computing systems;
- the use of parallel computing to solve complex problems in science and technology.
With our teaching programs we aim to provide the students with sound know how, that is close to research, in the field of Parallel Computing, with comprised skills for practical use and with experiences in real projects.