Using machine learning for data center cooling infrastructure efficiency prediction

Hayk Shoukourian; Torsten Wilde; Detlef Labrenz; Arndt Bode

doi:10.1109/IPDPSW.2017.25

Using machine learning for data center cooling infrastructure efficiency prediction

Hayk Shoukourian
, Torsten Wilde
, Detlef Labrenz
, Arndt Bode

Leibniz Rechenzentrum München

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

20 Scopus citations

Abstract

Power consumption continues to remain a critical aspect for High Performance Computing (HPC) data centers. It becomes even more crucial for Exascale computing since scaling today's fastest system to an Exaflop level would consume more than 168 MW power which is 8 times higher than the 20 MW power consumption goal set, at the time of this publication, by the US Department of Energy. This naturally leads to a necessity for energy efficiency improvement that will encompass the full chain of the power consumers, starting from the data center infrastructure, including cooling overheads and electrical losses, up to compute resource scheduling and application scaling. In this paper a machine learning approach is proposed to model the Coefficient of Performance (COP) of HPC data center's hot water cooling loop. The suggested model is validated on operational data obtained at Leibniz Supercomputing Centre (LRZ). The paper shows how this COP model can help to improve the energy efficiency of modern HPC data centers.

Original language	English
Title of host publication	Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	954-963
Number of pages	10
ISBN (Electronic)	9781538634080
DOIs	https://doi.org/10.1109/IPDPSW.2017.25
State	Published - 30 Jun 2017
Externally published	Yes
Event	31st IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017 - Orlando, United States Duration: 29 May 2017 → 2 Jun 2017

Publication series

Name	Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017

Conference

Conference	31st IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017
Country/Territory	United States
City	Orlando
Period	29/05/17 → 2/06/17

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

COP
data center
energy efficiency
high performance computing
machine learning

Access to Document

10.1109/IPDPSW.2017.25

Cite this

Shoukourian, H., Wilde, T., Labrenz, D., & Bode, A. (2017). Using machine learning for data center cooling infrastructure efficiency prediction. In Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017 (pp. 954-963). Article 7965143 (Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IPDPSW.2017.25

Shoukourian, Hayk ; Wilde, Torsten ; Labrenz, Detlef et al. / Using machine learning for data center cooling infrastructure efficiency prediction. Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 954-963 (Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017).

@inproceedings{f9e523415281474b9479ca368ffe3b5b,

title = "Using machine learning for data center cooling infrastructure efficiency prediction",

abstract = "Power consumption continues to remain a critical aspect for High Performance Computing (HPC) data centers. It becomes even more crucial for Exascale computing since scaling today's fastest system to an Exaflop level would consume more than 168 MW power which is 8 times higher than the 20 MW power consumption goal set, at the time of this publication, by the US Department of Energy. This naturally leads to a necessity for energy efficiency improvement that will encompass the full chain of the power consumers, starting from the data center infrastructure, including cooling overheads and electrical losses, up to compute resource scheduling and application scaling. In this paper a machine learning approach is proposed to model the Coefficient of Performance (COP) of HPC data center's hot water cooling loop. The suggested model is validated on operational data obtained at Leibniz Supercomputing Centre (LRZ). The paper shows how this COP model can help to improve the energy efficiency of modern HPC data centers.",

keywords = "COP, data center, energy efficiency, high performance computing, machine learning",

author = "Hayk Shoukourian and Torsten Wilde and Detlef Labrenz and Arndt Bode",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 31st IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017 ; Conference date: 29-05-2017 Through 02-06-2017",

year = "2017",

month = jun,

day = "30",

doi = "10.1109/IPDPSW.2017.25",

language = "English",

series = "Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017",

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

pages = "954--963",

booktitle = "Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017",

}

Shoukourian, H, Wilde, T, Labrenz, D & Bode, A 2017, Using machine learning for data center cooling infrastructure efficiency prediction. in Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017., 7965143, Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017, Institute of Electrical and Electronics Engineers Inc., pp. 954-963, 31st IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017, Orlando, United States, 29/05/17. https://doi.org/10.1109/IPDPSW.2017.25

Using machine learning for data center cooling infrastructure efficiency prediction. / Shoukourian, Hayk; Wilde, Torsten; Labrenz, Detlef et al.
Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 954-963 7965143 (Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017).

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

TY - GEN

T1 - Using machine learning for data center cooling infrastructure efficiency prediction

AU - Shoukourian, Hayk

AU - Wilde, Torsten

AU - Labrenz, Detlef

AU - Bode, Arndt

PY - 2017/6/30

Y1 - 2017/6/30

N2 - Power consumption continues to remain a critical aspect for High Performance Computing (HPC) data centers. It becomes even more crucial for Exascale computing since scaling today's fastest system to an Exaflop level would consume more than 168 MW power which is 8 times higher than the 20 MW power consumption goal set, at the time of this publication, by the US Department of Energy. This naturally leads to a necessity for energy efficiency improvement that will encompass the full chain of the power consumers, starting from the data center infrastructure, including cooling overheads and electrical losses, up to compute resource scheduling and application scaling. In this paper a machine learning approach is proposed to model the Coefficient of Performance (COP) of HPC data center's hot water cooling loop. The suggested model is validated on operational data obtained at Leibniz Supercomputing Centre (LRZ). The paper shows how this COP model can help to improve the energy efficiency of modern HPC data centers.

AB - Power consumption continues to remain a critical aspect for High Performance Computing (HPC) data centers. It becomes even more crucial for Exascale computing since scaling today's fastest system to an Exaflop level would consume more than 168 MW power which is 8 times higher than the 20 MW power consumption goal set, at the time of this publication, by the US Department of Energy. This naturally leads to a necessity for energy efficiency improvement that will encompass the full chain of the power consumers, starting from the data center infrastructure, including cooling overheads and electrical losses, up to compute resource scheduling and application scaling. In this paper a machine learning approach is proposed to model the Coefficient of Performance (COP) of HPC data center's hot water cooling loop. The suggested model is validated on operational data obtained at Leibniz Supercomputing Centre (LRZ). The paper shows how this COP model can help to improve the energy efficiency of modern HPC data centers.

KW - COP

KW - data center

KW - energy efficiency

KW - high performance computing

KW - machine learning

UR - https://www.scopus.com/pages/publications/85028046079

U2 - 10.1109/IPDPSW.2017.25

DO - 10.1109/IPDPSW.2017.25

M3 - Conference contribution

AN - SCOPUS:85028046079

T3 - Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017

SP - 954

EP - 963

BT - Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 31st IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017

Y2 - 29 May 2017 through 2 June 2017

ER -

Shoukourian H, Wilde T, Labrenz D, Bode A. Using machine learning for data center cooling infrastructure efficiency prediction. In Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 954-963. 7965143. (Proceedings - 2017 IEEE 31st International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2017). doi: 10.1109/IPDPSW.2017.25

Using machine learning for data center cooling infrastructure efficiency prediction

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this