Where academic tradition
meets the exciting future

Actor Merging for Dataflow Process Networks

Jani Boutellier, Johan Ersfolk, Johan Lilius, Marco Mattavelli, Ghislain Roquier, Olli Silven, Actor Merging for Dataflow Process Networks. IEEE Transactions on Signal Processing 63(10), 2496–2508, 2015.

http://dx.doi.org/10.1109/TSP.2015.2411229

Abstract:

Dataflow process networks provide a versatile model of computation for specifying signal processing applications in a platform independent fashion. This attractive feature of dataflow has lately been realized in dataflow programming tools that allow synthesizing the same application specification as both fixed hardware circuits and as software for programmable processors. However, in practice, the specification granularity of the dataflow program remains an arbitrary choice of the designer. Dataflow specifications of the same application with equivalent I/O behaviour can range from a single dataflow actor to a very fine grained network composed of elementary processing operations. A very fine grained dataflow specification might result into a high performance implementation when synthesized as hardware, but might perform poorly when executed on a programmable processor. This article presents actor merging as one solution for this performance portability problem of dataflow programs. In contrast to previous work around actor merging, this article presents a methodology that can merge also dynamic dataflow actors. To support these claims, results of experiments on several processing platforms and application examples ranging from telecommunications to video compression are reported.

BibTeX entry:

@ARTICLE{jBoErLiMaRoSi15a,
  title = {Actor Merging for Dataflow Process Networks},
  author = {Boutellier, Jani and Ersfolk, Johan and Lilius, Johan and Mattavelli, Marco and Roquier, Ghislain and Silven, Olli},
  journal = {IEEE Transactions on Signal Processing},
  volume = {63},
  number = {10},
  publisher = {IEEE},
  pages = {2496–2508},
  year = {2015},
  keywords = {Computational modeling;Hardware;Materials;Merging;Program processors;Signal processing;Dataflow computing;design automation;signal processing},
  ISSN = {1053-587X},
}

Belongs to TUCS Research Unit(s): Embedded Systems Laboratory (ESLAB)

Publication Forum rating of this publication: level 3

Edit publication