Task-based approaches are popular for the development of parallel programs for several reasons. They provide a decoupling of the parallel specification from the scheduling and mapping to the execution resources of a specific hardware platform, thus allowing a flexible and individual mapping. For platforms with a distributed address space, the use of parallel tasks instead of sequential tasks adds the advantage of structuring the program into communication domains that can help to reduce the overall communication overhead. In this talk, we consider the parallel programming model of communicating parallel tasks (CM-tasks), which allows both task-internal communication as well as communication between concurrently executed tasks at arbitrary points of their execution. We propose a corresponding scheduling algorithm and describe how the scheduling is supported by a transformation tool. An experimental evaluation of several application programs shows that using the CM-task model may lead to significant performance improvements compared to other parallel execution schemes.

About the Speaker: In 2002 Thomas Rauber joined the University of Bayreuth, where he now holds the chair for parallel and distributed computing. His research interest is focused on parallel processing, with an emphasis on task-based programming approaches and related scheduling and load balancing techniques. Other research interests include scalability issues for parallel numerical algorithms and program development for cloud computing. More information about his research can be found at ai2.uni-bayreuth.de. Dr. Rauber received a Ph.D. degree in computer science in 1990 from Saarland University in Saarbrücken and a habilitation degree in 1996 from the same university. From 1996 to 2002 he was Professor of Computer Science at the University Halle-Wittenberg.