TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

PRUNE: Dynamic and Decidable Dataflow for Signal Processing on Heterogeneous Platforms

Tutkimustuotosvertaisarvioitu

Standard

PRUNE: Dynamic and Decidable Dataflow for Signal Processing on Heterogeneous Platforms. / Boutellier, Jani; Wu, Jiahao; Huttunen, Heikki; Bhattacharyya, Shuvra.

julkaisussa: IEEE Transactions on Signal Processing, Vuosikerta 66, Nro 3, 2018, s. 654-665.

Tutkimustuotosvertaisarvioitu

Harvard

Boutellier, J, Wu, J, Huttunen, H & Bhattacharyya, S 2018, 'PRUNE: Dynamic and Decidable Dataflow for Signal Processing on Heterogeneous Platforms', IEEE Transactions on Signal Processing, Vuosikerta. 66, Nro 3, Sivut 654-665. https://doi.org/10.1109/TSP.2017.2773424

APA

Boutellier, J., Wu, J., Huttunen, H., & Bhattacharyya, S. (2018). PRUNE: Dynamic and Decidable Dataflow for Signal Processing on Heterogeneous Platforms. IEEE Transactions on Signal Processing, 66(3), 654-665. https://doi.org/10.1109/TSP.2017.2773424

Vancouver

Boutellier J, Wu J, Huttunen H, Bhattacharyya S. PRUNE: Dynamic and Decidable Dataflow for Signal Processing on Heterogeneous Platforms. IEEE Transactions on Signal Processing. 2018;66(3):654-665. https://doi.org/10.1109/TSP.2017.2773424

Author

Boutellier, Jani ; Wu, Jiahao ; Huttunen, Heikki ; Bhattacharyya, Shuvra. / PRUNE: Dynamic and Decidable Dataflow for Signal Processing on Heterogeneous Platforms. Julkaisussa: IEEE Transactions on Signal Processing. 2018 ; Vuosikerta 66, Nro 3. Sivut 654-665.

Bibtex - Lataa

@article{037a4bcb842741d5955e114c52e3364f,
title = "PRUNE: Dynamic and Decidable Dataflow for Signal Processing on Heterogeneous Platforms",
abstract = "The majority of contemporary mobile devices and personal computers are based on heterogeneous computing platforms that consist of a number of CPU cores and one or more Graphics Processing Units (GPUs). Despite the high volume of these devices, there are few existing programming frameworks that target full and simultaneous utilization of all CPU and GPU devices of the platform.This article presents a dataflow-flavored Model of Computation (MoC) that has been developed for deploying signal processing applications to heterogeneous platforms. The presented MoC is dynamic and allows describing applications with data dependent run-time behavior. On top of the MoC, formal design rules are presented that enable application descriptions to be simultaneously dynamic and decidable. Decidability guarantees compile-time application analyzability for deadlock freedom and bounded memory.The presented MoC and the design rules are realized in a novel Open Source programming environment {"}PRUNE'' and demonstrated with representative application examples from the domains of image processing, computer vision and wireless communications. Experimental results show that the proposed approach outperforms the state-of-the-art in analyzability, flexibility and performance.",
keywords = "Dataflow computing, design automation, signal processing, parallel processing, STREAMING APPLICATIONS, PROCESS NETWORKS, SYSTEMS, GRAPHS",
author = "Jani Boutellier and Jiahao Wu and Heikki Huttunen and Shuvra Bhattacharyya",
year = "2018",
doi = "10.1109/TSP.2017.2773424",
language = "English",
volume = "66",
pages = "654--665",
journal = "IEEE Transactions on Signal Processing",
issn = "1053-587X",
publisher = "IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC",
number = "3",

}

RIS (suitable for import to EndNote) - Lataa

TY - JOUR

T1 - PRUNE: Dynamic and Decidable Dataflow for Signal Processing on Heterogeneous Platforms

AU - Boutellier, Jani

AU - Wu, Jiahao

AU - Huttunen, Heikki

AU - Bhattacharyya, Shuvra

PY - 2018

Y1 - 2018

N2 - The majority of contemporary mobile devices and personal computers are based on heterogeneous computing platforms that consist of a number of CPU cores and one or more Graphics Processing Units (GPUs). Despite the high volume of these devices, there are few existing programming frameworks that target full and simultaneous utilization of all CPU and GPU devices of the platform.This article presents a dataflow-flavored Model of Computation (MoC) that has been developed for deploying signal processing applications to heterogeneous platforms. The presented MoC is dynamic and allows describing applications with data dependent run-time behavior. On top of the MoC, formal design rules are presented that enable application descriptions to be simultaneously dynamic and decidable. Decidability guarantees compile-time application analyzability for deadlock freedom and bounded memory.The presented MoC and the design rules are realized in a novel Open Source programming environment "PRUNE'' and demonstrated with representative application examples from the domains of image processing, computer vision and wireless communications. Experimental results show that the proposed approach outperforms the state-of-the-art in analyzability, flexibility and performance.

AB - The majority of contemporary mobile devices and personal computers are based on heterogeneous computing platforms that consist of a number of CPU cores and one or more Graphics Processing Units (GPUs). Despite the high volume of these devices, there are few existing programming frameworks that target full and simultaneous utilization of all CPU and GPU devices of the platform.This article presents a dataflow-flavored Model of Computation (MoC) that has been developed for deploying signal processing applications to heterogeneous platforms. The presented MoC is dynamic and allows describing applications with data dependent run-time behavior. On top of the MoC, formal design rules are presented that enable application descriptions to be simultaneously dynamic and decidable. Decidability guarantees compile-time application analyzability for deadlock freedom and bounded memory.The presented MoC and the design rules are realized in a novel Open Source programming environment "PRUNE'' and demonstrated with representative application examples from the domains of image processing, computer vision and wireless communications. Experimental results show that the proposed approach outperforms the state-of-the-art in analyzability, flexibility and performance.

KW - Dataflow computing

KW - design automation

KW - signal processing

KW - parallel processing

KW - STREAMING APPLICATIONS

KW - PROCESS NETWORKS

KW - SYSTEMS

KW - GRAPHS

UR - https://gitlab.com/jboutell/Prune

U2 - 10.1109/TSP.2017.2773424

DO - 10.1109/TSP.2017.2773424

M3 - Article

VL - 66

SP - 654

EP - 665

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

SN - 1053-587X

IS - 3

ER -