Shared-variable Synchronization Approaches for Dynamic Data Flow Programs

This paper presents shared-variable synchronization approaches for dataflow programming. The mechanisms do not require any substantial model of computation (MoC) modification, and is portable across both for hardware (HW) and software (SW) low-level code synthesis. With the shared-variable formalization, the benefits of the dataflow MoC are maintained, however the space and energy efficiency of an application can be significantly improved. The approach targets Dynamic Process Network (DPN) dataflow applications, thus making them also suitable for less expressive models e. g. synchronous and cyclostatic dataflow that DPN subsumes. The approach is validated through the analysis and optimization of a High-Efficiency Video Coding (HEVC) decoder implemented in the RVC-CAL dataflow language targeting a multi-core platform. Experimental results show how, starting from an initial design that does not use the shared-variable formalism, frames per second throughput performance is increased by a factor of 21.


Publié dans:
Proceedings Of The 2018 Ieee International Workshop On Signal Processing Systems (Sips), 263-268
Présenté à:
IEEE International Workshop on Signal Processing Systems (IEEE SiPS), Cape Town, SOUTH AFRICA, Oct 21-24, 2018
Année
Jan 01 2018
Publisher:
New York, IEEE
ISSN:
1520-6130
ISBN:
978-1-5386-6318-9
Laboratoires:




 Notice créée le 2019-06-18, modifiée le 2019-08-12


Évaluer ce document:

Rate this document:
1
2
3
 
(Pas encore évalué)