A machine model for dataflow actors and its applications

In application areas that process stream-like data such as multimedia, networking and DSP, the pipelined concurrent processing is frequently represented as a dataflow network of communicating computational kernels connected by FIFO queues. However, while dataflow is a natural medium for conceptualizing and modeling stream-processing systems, its adoption as a programming methodology has been hindered by an unappealing choice between expressiveness and efficient implementability—efficient implementation techniques being primarily limited to restricted subclasses of dataflow programs. The paper presents a simple machine model for a very general class of dataflow programs and shows how it can be used as a foundation for their efficient implementation.

[1]  E.A. Lee,et al.  Synchronous data flow , 1987, Proceedings of the IEEE.

[2]  Mickaël Raulet,et al.  Exploiting Statically Schedulable Regions in Dataflow Programs , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.

[3]  Scalable parallelism using dataflow programming , 2010 .

[4]  Jörn W. Janneck,et al.  Actors and their Composition , 2003, Formal Aspects of Computing.

[5]  Gilles Kahn,et al.  The Semantics of a Simple Language for Parallel Programming , 1974, IFIP Congress.

[6]  Ghislain Roquier,et al.  Software Code Generation for the RVC-CAL Language , 2011, J. Signal Process. Syst..

[7]  Rudy Lauwereins,et al.  Static scheduling of multi-rate and cyclo-static DSP-applications , 1994, Proceedings of 1994 IEEE Workshop on VLSI Signal Processing.

[8]  Jack B. Dennis,et al.  First version of a data flow procedure language , 1974, Symposium on Programming.

[9]  Mickaël Raulet,et al.  Overview of the MPEG Reconfigurable Video Coding Framework , 2011, J. Signal Process. Syst..

[10]  Edward A. Lee,et al.  Dataflow process networks , 1995, Proc. IEEE.