University of Oulu

Hautala, I., Boutellier, J., Nyländen, T. et al. J Sign Process Syst (2018) 90: 1507. https://doi.org/10.1007/s11265-018-1339-x

Toward efficient execution of RVC-CAL dataflow programs on multicore platforms

Saved in:
Author: Hautala, Ilkka1; Boutellier, Jani2; Nyländen, Teemu1;
Organizations: 1Department of Computer Science and Engineering, University of Oulu, Oulu, Finland
2Department of Pervasive Computing, Tampere University of Technology, Tampere, Finland
Format: article
Version: accepted version
Access: open
Online Access: PDF Full Text (PDF, 0.2 MB)
Persistent link: http://urn.fi/urn:nbn:fi-fe2019071023003
Language: English
Published: Springer Nature, 2018
Publish Date: 2019-02-09
Description:

Abstract

The increasing number of cores in System on Chips (SoC) has introduced challenges in software parallelization. As an answer to this, the dataflow programming model offers a concurrent and reusability promoting approach for describing applications. In this work, a runtime for executing Dataflow Process Networks (DPN) on multicore platforms is proposed. The main difference between this work and existing methods is letting the operating system perform Central processing unit (CPU) load-balancing freely, instead of limiting thread migration between processing cores through CPU affinity. The proposed runtime is benchmarked on desktop and server multicore platforms using five different applications from video coding and telecommunication domains. The results show that the proposed method offers significant improvements over the state-of-art, in terms of performance and reliability.

see all

Series: Journal of signal processing systems for signal image and video technology
ISSN: 1939-8018
ISSN-E: 1939-8115
ISSN-L: 1939-8018
Volume: 90
Issue: 11
Pages: 1507 - 1517
DOI: 10.1007/s11265-018-1339-x
OADOI: https://oadoi.org/10.1007/s11265-018-1339-x
Type of Publication: A1 Journal article – refereed
Field of Science: 113 Computer and information sciences
Subjects:
Copyright information: © Springer Science+Business Media, LLC, part of Springer Nature 2018. This is a post-peer-review, pre-copyedit version of an article published in Journal of Signal Processing Systems. The final authenticated version is available online at: https://doi.org/10.1007/s11265-018-1339-x.