A reinforcement learning-based power management framework for green computing data centers

Xue Lin; Yanzhi Wang; Massoud Pedram

doi:10.1109/IC2E.2016.33

A reinforcement learning-based power management framework for green computing data centers

Xue Lin, Yanzhi Wang, Massoud Pedram

Department of Electrical Engineering & Computer Science

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

18 Scopus citations

Abstract

Various power management techniques have been exploited to reduce the energy consumption of data centers. In this work, we propose a reinforcement learning-based power management framework for data centers, which does not rely on any given stationary assumptions of the job arrival and job service processes. By carefully designing the state space, the action space, and the reward of a learning process, the objective of the reinforcement learning agent coincides with our goal of reducing the server pool energy consumption with reasonable average job response time. Real Google cluster data traces are used to verify the effectiveness of the proposed reinforcement learning-based data center power management framework.

Original language	English (US)
Title of host publication	Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016
Subtitle of host publication	Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	135-138
Number of pages	4
ISBN (Electronic)	9781509019618
DOIs	https://doi.org/10.1109/IC2E.2016.33
State	Published - Jun 1 2016
Event	4th IEEE Annual International Conference on Cloud Engineering, IC2E 2016 - Berlin, Germany Duration: Apr 4 2016 → Apr 8 2016

Publication series

Name	Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016

Other

Other	4th IEEE Annual International Conference on Cloud Engineering, IC2E 2016
Country	Germany
City	Berlin
Period	4/4/16 → 4/8/16

Keywords

power management
reinforcement learning

ASJC Scopus subject areas

Control and Systems Engineering
Computer Networks and Communications

Access to Document

10.1109/IC2E.2016.33

Fingerprint Dive into the research topics of 'A reinforcement learning-based power management framework for green computing data centers'. Together they form a unique fingerprint.

Cite this

Lin, X., Wang, Y., & Pedram, M. (2016). A reinforcement learning-based power management framework for green computing data centers. In Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016 (pp. 135-138). [7484175] (Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IC2E.2016.33

A reinforcement learning-based power management framework for green computing data centers. / Lin, Xue; Wang, Yanzhi; Pedram, Massoud.

Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016. Institute of Electrical and Electronics Engineers Inc., 2016. p. 135-138 7484175 (Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016).

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

Lin, X, Wang, Y & Pedram, M 2016, A reinforcement learning-based power management framework for green computing data centers. in Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016., 7484175, Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016, Institute of Electrical and Electronics Engineers Inc., pp. 135-138, 4th IEEE Annual International Conference on Cloud Engineering, IC2E 2016, Berlin, Germany, 4/4/16. https://doi.org/10.1109/IC2E.2016.33

Lin X, Wang Y, Pedram M. A reinforcement learning-based power management framework for green computing data centers. In Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 135-138. 7484175. (Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016). https://doi.org/10.1109/IC2E.2016.33

Lin, Xue ; Wang, Yanzhi ; Pedram, Massoud. / A reinforcement learning-based power management framework for green computing data centers. Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 135-138 (Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016).

@inproceedings{fd583773c5c24bc1ac221f1276f38534,

title = "A reinforcement learning-based power management framework for green computing data centers",

abstract = "Various power management techniques have been exploited to reduce the energy consumption of data centers. In this work, we propose a reinforcement learning-based power management framework for data centers, which does not rely on any given stationary assumptions of the job arrival and job service processes. By carefully designing the state space, the action space, and the reward of a learning process, the objective of the reinforcement learning agent coincides with our goal of reducing the server pool energy consumption with reasonable average job response time. Real Google cluster data traces are used to verify the effectiveness of the proposed reinforcement learning-based data center power management framework.",

keywords = "power management, reinforcement learning",

author = "Xue Lin and Yanzhi Wang and Massoud Pedram",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE. Copyright: Copyright 2016 Elsevier B.V., All rights reserved.; 4th IEEE Annual International Conference on Cloud Engineering, IC2E 2016 ; Conference date: 04-04-2016 Through 08-04-2016",

year = "2016",

month = jun,

day = "1",

doi = "10.1109/IC2E.2016.33",

language = "English (US)",

series = "Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016",

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

pages = "135--138",

booktitle = "Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016",

}

TY - GEN

T1 - A reinforcement learning-based power management framework for green computing data centers

AU - Lin, Xue

AU - Wang, Yanzhi

AU - Pedram, Massoud

PY - 2016/6/1

Y1 - 2016/6/1

N2 - Various power management techniques have been exploited to reduce the energy consumption of data centers. In this work, we propose a reinforcement learning-based power management framework for data centers, which does not rely on any given stationary assumptions of the job arrival and job service processes. By carefully designing the state space, the action space, and the reward of a learning process, the objective of the reinforcement learning agent coincides with our goal of reducing the server pool energy consumption with reasonable average job response time. Real Google cluster data traces are used to verify the effectiveness of the proposed reinforcement learning-based data center power management framework.

AB - Various power management techniques have been exploited to reduce the energy consumption of data centers. In this work, we propose a reinforcement learning-based power management framework for data centers, which does not rely on any given stationary assumptions of the job arrival and job service processes. By carefully designing the state space, the action space, and the reward of a learning process, the objective of the reinforcement learning agent coincides with our goal of reducing the server pool energy consumption with reasonable average job response time. Real Google cluster data traces are used to verify the effectiveness of the proposed reinforcement learning-based data center power management framework.

KW - power management

KW - reinforcement learning

UR - http://www.scopus.com/inward/record.url?scp=84978079861&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84978079861&partnerID=8YFLogxK

U2 - 10.1109/IC2E.2016.33

DO - 10.1109/IC2E.2016.33

M3 - Conference contribution

AN - SCOPUS:84978079861

T3 - Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016: Co-located with the 1st IEEE International Conference on Internet-of-Things Design and Implementation, IoTDI 2016

SP - 135

EP - 138

BT - Proceedings - 2016 IEEE International Conference on Cloud Engineering, IC2E 2016

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 4th IEEE Annual International Conference on Cloud Engineering, IC2E 2016

Y2 - 4 April 2016 through 8 April 2016

ER -