Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs

David Akselrod

doi:10.1109/SOCC52499.2021.9739607

Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs

David Akselrod

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

This paper presents a Power/Performance-optimal Markov Decision Processes (MDP) based DVFS (Dynamic Voltage and Frequency Scaling) controller. MDP framework is employed for DVFS switching decision-making under conditions of uncertainly. The approach yields an optimal Power/Performance DVFS technique based on provided reward/performance structure, presenting a dynamically adjusted DVFS mechanism in respect to the formulated model.

Original language	English
Title of host publication	Proceedings - 34th IEEE International System-on-Chip Conference, SOCC 2021
Editors	Gang Qu, Jinjun Xiong, Danella Zhao, Venki Muthukumar, Md Farhadur Reza, Ramalingam Sridhar
Publisher	Institute of Electrical and Electronics Engineers
Pages	71-74
Number of pages	4
ISBN (Electronic)	9781665429313
DOIs	https://doi.org/10.1109/SOCC52499.2021.9739607
State	Published - 1 Jan 2021
Externally published	Yes
Event	34th IEEE International System-on-Chip Conference, SOCC 2021 - Virtual, Online, United States Duration: 14 Sep 2021 → 17 Sep 2021

Publication series

Name	International System on Chip Conference
Volume	2021-September
ISSN (Print)	2164-1676
ISSN (Electronic)	2164-1706

Conference

Conference	34th IEEE International System-on-Chip Conference, SOCC 2021
Country/Territory	United States
City	Virtual, Online
Period	14/09/21 → 17/09/21

ASJC Scopus subject areas

Hardware and Architecture
Control and Systems Engineering
Electrical and Electronic Engineering

Access to Document

10.1109/SOCC52499.2021.9739607

Cite this

Akselrod, D. (2021). Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs. In G. Qu, J. Xiong, D. Zhao, V. Muthukumar, M. F. Reza, & R. Sridhar (Eds.), Proceedings - 34th IEEE International System-on-Chip Conference, SOCC 2021 (pp. 71-74). (International System on Chip Conference; Vol. 2021-September). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/SOCC52499.2021.9739607

Akselrod, David. / Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs. Proceedings - 34th IEEE International System-on-Chip Conference, SOCC 2021. editor / Gang Qu ; Jinjun Xiong ; Danella Zhao ; Venki Muthukumar ; Md Farhadur Reza ; Ramalingam Sridhar. Institute of Electrical and Electronics Engineers, 2021. pp. 71-74 (International System on Chip Conference).

@inproceedings{101760b7bdef43c98a88ab77e84c07bc,

title = "Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs",

abstract = "This paper presents a Power/Performance-optimal Markov Decision Processes (MDP) based DVFS (Dynamic Voltage and Frequency Scaling) controller. MDP framework is employed for DVFS switching decision-making under conditions of uncertainly. The approach yields an optimal Power/Performance DVFS technique based on provided reward/performance structure, presenting a dynamically adjusted DVFS mechanism in respect to the formulated model.",

author = "David Akselrod",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 34th IEEE International System-on-Chip Conference, SOCC 2021 ; Conference date: 14-09-2021 Through 17-09-2021",

year = "2021",

month = jan,

day = "1",

doi = "10.1109/SOCC52499.2021.9739607",

language = "English",

series = "International System on Chip Conference",

publisher = "Institute of Electrical and Electronics Engineers",

pages = "71--74",

editor = "Gang Qu and Jinjun Xiong and Danella Zhao and Venki Muthukumar and Reza, {Md Farhadur} and Ramalingam Sridhar",

booktitle = "Proceedings - 34th IEEE International System-on-Chip Conference, SOCC 2021",

address = "United States",

}

Akselrod, D 2021, Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs. in G Qu, J Xiong, D Zhao, V Muthukumar, MF Reza & R Sridhar (eds), Proceedings - 34th IEEE International System-on-Chip Conference, SOCC 2021. International System on Chip Conference, vol. 2021-September, Institute of Electrical and Electronics Engineers, pp. 71-74, 34th IEEE International System-on-Chip Conference, SOCC 2021, Virtual, Online, United States, 14/09/21. https://doi.org/10.1109/SOCC52499.2021.9739607

Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs. / Akselrod, David.
Proceedings - 34th IEEE International System-on-Chip Conference, SOCC 2021. ed. / Gang Qu; Jinjun Xiong; Danella Zhao; Venki Muthukumar; Md Farhadur Reza; Ramalingam Sridhar. Institute of Electrical and Electronics Engineers, 2021. p. 71-74 (International System on Chip Conference; Vol. 2021-September).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs

AU - Akselrod, David

PY - 2021/1/1

Y1 - 2021/1/1

N2 - This paper presents a Power/Performance-optimal Markov Decision Processes (MDP) based DVFS (Dynamic Voltage and Frequency Scaling) controller. MDP framework is employed for DVFS switching decision-making under conditions of uncertainly. The approach yields an optimal Power/Performance DVFS technique based on provided reward/performance structure, presenting a dynamically adjusted DVFS mechanism in respect to the formulated model.

AB - This paper presents a Power/Performance-optimal Markov Decision Processes (MDP) based DVFS (Dynamic Voltage and Frequency Scaling) controller. MDP framework is employed for DVFS switching decision-making under conditions of uncertainly. The approach yields an optimal Power/Performance DVFS technique based on provided reward/performance structure, presenting a dynamically adjusted DVFS mechanism in respect to the formulated model.

UR - http://www.scopus.com/inward/record.url?scp=85127786137&partnerID=8YFLogxK

U2 - 10.1109/SOCC52499.2021.9739607

DO - 10.1109/SOCC52499.2021.9739607

M3 - Conference contribution

AN - SCOPUS:85127786137

T3 - International System on Chip Conference

SP - 71

EP - 74

BT - Proceedings - 34th IEEE International System-on-Chip Conference, SOCC 2021

A2 - Qu, Gang

A2 - Xiong, Jinjun

A2 - Zhao, Danella

A2 - Muthukumar, Venki

A2 - Reza, Md Farhadur

A2 - Sridhar, Ramalingam

PB - Institute of Electrical and Electronics Engineers

T2 - 34th IEEE International System-on-Chip Conference, SOCC 2021

Y2 - 14 September 2021 through 17 September 2021

ER -

Akselrod D. Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs. In Qu G, Xiong J, Zhao D, Muthukumar V, Reza MF, Sridhar R, editors, Proceedings - 34th IEEE International System-on-Chip Conference, SOCC 2021. Institute of Electrical and Electronics Engineers. 2021. p. 71-74. (International System on Chip Conference). doi: 10.1109/SOCC52499.2021.9739607

Reinforcement Learning-based Power Management Architecture for Optimal DVFS in SoCs

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this