Kees van Berkel

Technical University Eindhoven, The Netherlands

StaccatoLab: A Programming and Execution Model for Large-Scale Dataflow Computing

Abstract

Dataflow is explored as a promising programming model for large-scale parallel computing, both for embedded and for high-performance computing. With a suitable execution model, such as StaccatoLab, dataflow programs support reasoning about quantitative properties of programs, including throughput. For dataflow synthesis we assume FPGAs as a target.

Biography

Kees van Berkel started his R&D career at Philips Research in 1980, after receiving an MSc degree in EE from TU Delft. Since 2000 he has been a fellow at Philips, NXP, ST-Ericsson, and Ericsson, until mid 2015.  He obtained a PhD in CS from TU Eindhoven in 1992, where he is a part-time full professor since 1996. Since 2019 he is also part-time advisor to Grai Matter Labs.

Kees pioneered asynchronous VLSI from theory to mass production, as well as embedded vector processing for software-defined radio. His research interests include parallel programming, multi-core hardware architectures, resource management, low power computing, and deep learning.  His current research focus is on exascale computing for radio astronomy. See also http://www.win.tue.nl/~cberkel/StaccatoLab-program.html