TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

Toward Efficient Execution of RVC-CAL Dataflow Programs on Multicore Platforms

Tutkimustuotosvertaisarvioitu

Standard

Toward Efficient Execution of RVC-CAL Dataflow Programs on Multicore Platforms. / Hautala, Ilkka; Boutellier, Jani; Nyländen, Teemu; Silvén, Olli.

julkaisussa: Journal of Signal Processing Systems, Vuosikerta 90, Nro 11, 11.2018, s. 1507-1517.

Tutkimustuotosvertaisarvioitu

Harvard

Hautala, I, Boutellier, J, Nyländen, T & Silvén, O 2018, 'Toward Efficient Execution of RVC-CAL Dataflow Programs on Multicore Platforms', Journal of Signal Processing Systems, Vuosikerta. 90, Nro 11, Sivut 1507-1517. https://doi.org/10.1007/s11265-018-1339-x

APA

Hautala, I., Boutellier, J., Nyländen, T., & Silvén, O. (2018). Toward Efficient Execution of RVC-CAL Dataflow Programs on Multicore Platforms. Journal of Signal Processing Systems, 90(11), 1507-1517. https://doi.org/10.1007/s11265-018-1339-x

Vancouver

Hautala I, Boutellier J, Nyländen T, Silvén O. Toward Efficient Execution of RVC-CAL Dataflow Programs on Multicore Platforms. Journal of Signal Processing Systems. 2018 marras;90(11):1507-1517. https://doi.org/10.1007/s11265-018-1339-x

Author

Hautala, Ilkka ; Boutellier, Jani ; Nyländen, Teemu ; Silvén, Olli. / Toward Efficient Execution of RVC-CAL Dataflow Programs on Multicore Platforms. Julkaisussa: Journal of Signal Processing Systems. 2018 ; Vuosikerta 90, Nro 11. Sivut 1507-1517.

Bibtex - Lataa

@article{1ee68681e9af4bcb8cc59821f40d44a8,
title = "Toward Efficient Execution of RVC-CAL Dataflow Programs on Multicore Platforms",
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.",
keywords = "Dataflow Process Networks, Multicore, Orcc, RVC-CAL",
author = "Ilkka Hautala and Jani Boutellier and Teemu Nyl{\"a}nden and Olli Silv{\'e}n",
year = "2018",
month = "11",
doi = "10.1007/s11265-018-1339-x",
language = "English",
volume = "90",
pages = "1507--1517",
journal = "Journal of Signal Processing Systems",
issn = "1939-8018",
publisher = "Springer Verlag",
number = "11",

}

RIS (suitable for import to EndNote) - Lataa

TY - JOUR

T1 - Toward Efficient Execution of RVC-CAL Dataflow Programs on Multicore Platforms

AU - Hautala, Ilkka

AU - Boutellier, Jani

AU - Nyländen, Teemu

AU - Silvén, Olli

PY - 2018/11

Y1 - 2018/11

N2 - 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.

AB - 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.

KW - Dataflow Process Networks

KW - Multicore

KW - Orcc

KW - RVC-CAL

U2 - 10.1007/s11265-018-1339-x

DO - 10.1007/s11265-018-1339-x

M3 - Article

VL - 90

SP - 1507

EP - 1517

JO - Journal of Signal Processing Systems

JF - Journal of Signal Processing Systems

SN - 1939-8018

IS - 11

ER -