Physical and Data-Driven Models for Edge Data Center Cooling System

Mikko Siltala; Rickard Brannvall; Jonas Gustafsson; Quan Zhou

doi:10.1109/SweDS51247.2020.9275588

Physical and Data-Driven Models for Edge Data Center Cooling System

Mikko Siltala, Rickard Brannvall^*, Jonas Gustafsson, Quan Zhou

^*Corresponding author for this work

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

5 Downloads (Pure)

Abstract

Edge data centers are expected to become prevalent providing low latency computing power for 5G mobile and IoT applications. This article develops two models for the complete cooling system of an edge data center: One model based on the laws of thermodynamics and one data-driven model based on LSTM neural networks. The models are validated against an actual edge data center experimental set-up showing root mean squared errors (RMSE) for most individual components below 1 °C over a simulation period of approximately 10 hours; which compares favourably to state-of-the-art models.

Original language	English
Title of host publication	2020 Swedish Workshop on Data Science, SweDS 2020
Publisher	IEEE
Number of pages	7
ISBN (Electronic)	9781728192048
DOIs	https://doi.org/10.1109/SweDS51247.2020.9275588
Publication status	Published - 29 Oct 2020
MoE publication type	A4 Article in a conference publication
Event	Swedish Workshop on Data Science - Lulea, Sweden Duration: 29 Oct 2020 → 30 Oct 2020

Workshop

Workshop	Swedish Workshop on Data Science
Abbreviated title	SweDS
Country	Sweden
City	Lulea
Period	29/10/2020 → 30/10/2020

Keywords

Cooling System
Data center
Edge
LSTM
Thermal Energy Storage

Access to Document

10.1109/SweDS51247.2020.9275588

ELEC_Siltala_etal_Physical-and-Data-Driven_SweDS20_acceptedauthormanuscriptAccepted author manuscript, 3.69 MB

Link to publication in Scopus

Fingerprint Dive into the research topics of 'Physical and Data-Driven Models for Edge Data Center Cooling System'. Together they form a unique fingerprint.

Cite this

@inproceedings{303e7db8a40e4000944183bfea2f1ac6,

title = "Physical and Data-Driven Models for Edge Data Center Cooling System",

abstract = "Edge data centers are expected to become prevalent providing low latency computing power for 5G mobile and IoT applications. This article develops two models for the complete cooling system of an edge data center: One model based on the laws of thermodynamics and one data-driven model based on LSTM neural networks. The models are validated against an actual edge data center experimental set-up showing root mean squared errors (RMSE) for most individual components below 1 °C over a simulation period of approximately 10 hours; which compares favourably to state-of-the-art models. ",

keywords = "Cooling System, Data center, Edge, LSTM, Thermal Energy Storage",

author = "Mikko Siltala and Rickard Brannvall and Jonas Gustafsson and Quan Zhou",

year = "2020",

month = oct,

day = "29",

doi = "10.1109/SweDS51247.2020.9275588",

language = "English",

booktitle = "2020 Swedish Workshop on Data Science, SweDS 2020",

publisher = "IEEE",

address = "United States",

note = "Swedish Workshop on Data Science, SweDS ; Conference date: 29-10-2020 Through 30-10-2020",

}

TY - GEN

T1 - Physical and Data-Driven Models for Edge Data Center Cooling System

AU - Siltala, Mikko

AU - Brannvall, Rickard

AU - Gustafsson, Jonas

AU - Zhou, Quan

PY - 2020/10/29

Y1 - 2020/10/29

N2 - Edge data centers are expected to become prevalent providing low latency computing power for 5G mobile and IoT applications. This article develops two models for the complete cooling system of an edge data center: One model based on the laws of thermodynamics and one data-driven model based on LSTM neural networks. The models are validated against an actual edge data center experimental set-up showing root mean squared errors (RMSE) for most individual components below 1 °C over a simulation period of approximately 10 hours; which compares favourably to state-of-the-art models.

AB - Edge data centers are expected to become prevalent providing low latency computing power for 5G mobile and IoT applications. This article develops two models for the complete cooling system of an edge data center: One model based on the laws of thermodynamics and one data-driven model based on LSTM neural networks. The models are validated against an actual edge data center experimental set-up showing root mean squared errors (RMSE) for most individual components below 1 °C over a simulation period of approximately 10 hours; which compares favourably to state-of-the-art models.

KW - Cooling System

KW - Data center

KW - Edge

KW - LSTM

KW - Thermal Energy Storage

UR - http://www.scopus.com/inward/record.url?scp=85099112779&partnerID=8YFLogxK

U2 - 10.1109/SweDS51247.2020.9275588

DO - 10.1109/SweDS51247.2020.9275588

M3 - Conference contribution

AN - SCOPUS:85099112779

BT - 2020 Swedish Workshop on Data Science, SweDS 2020

PB - IEEE

T2 - Swedish Workshop on Data Science

Y2 - 29 October 2020 through 30 October 2020

ER -