Vagelis Bebelis

Abstract - Boolean Parametric Data Flow

Dataflow programming models are well-suited to program many-core streaming applications. However, many streaming applications have a dynamic behavior. To capture this behavior, parametric dataflow models have been introduced over the years. Still, such models do not allow the topology of the dataflow graph to change at runtime, a feature that is also required to program modern streaming applications. To overcome these restrictions, we propose a new model of computation, the Boolean Parametric Data Flow (BPDF) model which combines integer parameters (to express dynamic rates) and boolean parameters (to express the activation and deactivation of communication channels). The major challenge with such dynamic models is to guarantee liveness and boundedness. We present static analyses which ensure statically the liveness and the boundedness of BDPF graphs. Finally, a dynamic model like BPDF creates new challenges when it comes to its We present a flexible approach to produce parallel schedules for BPDF applications on many core platforms.

Short bio

Vagelis Bebelis is a PhD student at INRIA Grenoble in collaboration with STMicroelectronics. He is working on models of computation and more specifically data flow models. Before his PhD, he finsihed a bachelor on electrical engineering at the University of Patras, Greece and a master on embedded systems from the same university. He carried out his master thesis in IMEC Leuven, Belgium working on power optimization of the front-end of 60GHz radio.