Where academic tradition
meets the exciting future

Energy-Efficient Actor Execution for SDF Application on Heterogeneous Architectures

Hergys Rexha, Lafond Sébastien, Karol Desnos, Energy-Efficient Actor Execution for SDF Application on Heterogeneous Architectures. In: Juan E Guerrero (Ed.), 26th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing, 486–493, IEEE, 2018.

Abstract:

Heterogeneous systems promise to improve per- formance and endurance of power constrained systems, by utilizing computing elements of different power and performance characteristics. Such systems provide the possibility to trade number and types of core with Dynamic Voltage and Frequency Scaling (DVFS) levels and core utilization rate to achieve op- timal energy efficiency. Therefore by making smart decisions on application scheduling and mapping we can exploit and maximize the benefits of using heterogeneous processors. At the same time, the application level of parallelism can conveniently be exposed by dataflow Models of Computation (MoCs). In this paper we show an energy efficient execution approach for heterogeneous architecture. We demonstrate the approach on a real-life streaming application modelled with Parameterized and Interfaced Synchronous Dataflow (PiSDF). The presented solution shows how to integrate our approach in the workflow of a dataflow application prototyping tool. The obtained results demonstrate that, by using an optimal scheduling and mapping, more than 30% of energy reduction can be achieved on a single actor level.

BibTeX entry:

@INPROCEEDINGS{inpReLaDe18a,
  title = {Energy-Efficient Actor Execution for SDF Application on Heterogeneous Architectures},
  booktitle = {26th Euromicro International Conference on Parallel, Distributed, and Network-Based Processing},
  author = {Rexha, Hergys and Lafond Sébastien and Desnos, Karol},
  editor = {Guerrero, Juan E},
  publisher = {IEEE},
  pages = {486–493},
  year = {2018},
}

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

Publication Forum rating of this publication: level 1

Edit publication