A data-traffic aware dynamic power management for general-purpose graphics processing units

Lih Yih Chiou; Chao Kai Yang; Che Pin Chang

doi:10.1109/ISCAS.2019.8702405

A data-traffic aware dynamic power management for general-purpose graphics processing units

Lih Yih Chiou, Chao Kai Yang, Che Pin Chang

電機工程學系

研究成果: Conference contribution

摘要

General-purpose graphics processing units (GPGPUs) are gaining popularity. Applications like machine learning and artificial intelligence have been growing rapidly. When hardware complexity increases to meet the computing requirements of the applications, the power efficiency of the hardware becomes an important issue. In this paper, we propose an adaptive power management scheme for GPGPU to enhance power efficiency by proactively detecting the internal activities during runtime and dynamically adjusting the operating frequencies of selected components. The simulation results indicate that, when compared with the baseline approach, the proposed scheme can reduce power consumption by up to 15% with a performance overhead of less than 5% on average.

原文	English
主出版物標題	2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019 - Proceedings
發行者	Institute of Electrical and Electronics Engineers Inc.
ISBN（電子）	9781728103976
DOIs	https://doi.org/10.1109/ISCAS.2019.8702405
出版狀態	Published - 2019 一月 1
事件	2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019 - Sapporo, Japan 持續時間: 2019 五月 26 → 2019 五月 29

出版系列

名字	Proceedings - IEEE International Symposium on Circuits and Systems
卷	2019-May
ISSN（列印）	0271-4310

Conference

Conference	2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019
國家	Japan
城市	Sapporo
期間	19-05-26 → 19-05-29

All Science Journal Classification (ASJC) codes

Electrical and Electronic Engineering

存取文件

10.1109/ISCAS.2019.8702405

指紋深入研究「A data-traffic aware dynamic power management for general-purpose graphics processing units」主題。共同形成了獨特的指紋。

引用此

Chiou, L. Y., Yang, C. K., & Chang, C. P. (2019). A data-traffic aware dynamic power management for general-purpose graphics processing units. 於 2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019 - Proceedings [8702405] (Proceedings - IEEE International Symposium on Circuits and Systems; 卷 2019-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISCAS.2019.8702405

@inproceedings{d50bf6196b5348daab49a4e8d30711b7,

title = "A data-traffic aware dynamic power management for general-purpose graphics processing units",

abstract = "General-purpose graphics processing units (GPGPUs) are gaining popularity. Applications like machine learning and artificial intelligence have been growing rapidly. When hardware complexity increases to meet the computing requirements of the applications, the power efficiency of the hardware becomes an important issue. In this paper, we propose an adaptive power management scheme for GPGPU to enhance power efficiency by proactively detecting the internal activities during runtime and dynamically adjusting the operating frequencies of selected components. The simulation results indicate that, when compared with the baseline approach, the proposed scheme can reduce power consumption by up to 15% with a performance overhead of less than 5% on average.",

author = "Chiou, {Lih Yih} and Yang, {Chao Kai} and Chang, {Che Pin}",

year = "2019",

month = jan,

day = "1",

doi = "10.1109/ISCAS.2019.8702405",

language = "English",

series = "Proceedings - IEEE International Symposium on Circuits and Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019 - Proceedings",

address = "United States",

note = "2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019 ; Conference date: 26-05-2019 Through 29-05-2019",

}

Chiou, LY, Yang, CK & Chang, CP 2019, A data-traffic aware dynamic power management for general-purpose graphics processing units. 於 2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019 - Proceedings., 8702405, Proceedings - IEEE International Symposium on Circuits and Systems, 卷 2019-May, Institute of Electrical and Electronics Engineers Inc., 2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019, Sapporo, Japan, 19-05-26. https://doi.org/10.1109/ISCAS.2019.8702405

A data-traffic aware dynamic power management for general-purpose graphics processing units. / Chiou, Lih Yih; Yang, Chao Kai; Chang, Che Pin.

2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. 8702405 (Proceedings - IEEE International Symposium on Circuits and Systems; 卷 2019-May).

研究成果: Conference contribution

TY - GEN

T1 - A data-traffic aware dynamic power management for general-purpose graphics processing units

AU - Chiou, Lih Yih

AU - Yang, Chao Kai

AU - Chang, Che Pin

PY - 2019/1/1

Y1 - 2019/1/1

N2 - General-purpose graphics processing units (GPGPUs) are gaining popularity. Applications like machine learning and artificial intelligence have been growing rapidly. When hardware complexity increases to meet the computing requirements of the applications, the power efficiency of the hardware becomes an important issue. In this paper, we propose an adaptive power management scheme for GPGPU to enhance power efficiency by proactively detecting the internal activities during runtime and dynamically adjusting the operating frequencies of selected components. The simulation results indicate that, when compared with the baseline approach, the proposed scheme can reduce power consumption by up to 15% with a performance overhead of less than 5% on average.

AB - General-purpose graphics processing units (GPGPUs) are gaining popularity. Applications like machine learning and artificial intelligence have been growing rapidly. When hardware complexity increases to meet the computing requirements of the applications, the power efficiency of the hardware becomes an important issue. In this paper, we propose an adaptive power management scheme for GPGPU to enhance power efficiency by proactively detecting the internal activities during runtime and dynamically adjusting the operating frequencies of selected components. The simulation results indicate that, when compared with the baseline approach, the proposed scheme can reduce power consumption by up to 15% with a performance overhead of less than 5% on average.

UR - http://www.scopus.com/inward/record.url?scp=85066780769&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85066780769&partnerID=8YFLogxK

U2 - 10.1109/ISCAS.2019.8702405

DO - 10.1109/ISCAS.2019.8702405

M3 - Conference contribution

AN - SCOPUS:85066780769

T3 - Proceedings - IEEE International Symposium on Circuits and Systems

BT - 2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019

Y2 - 26 May 2019 through 29 May 2019

ER -

Chiou LY, Yang CK, Chang CP. A data-traffic aware dynamic power management for general-purpose graphics processing units. 於 2019 IEEE International Symposium on Circuits and Systems, ISCAS 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. 8702405. (Proceedings - IEEE International Symposium on Circuits and Systems). https://doi.org/10.1109/ISCAS.2019.8702405