Energy efficient decision making in data centers with multiple cooling methods

Arash Mousavi, Yulia Berezovskaya, Valeriy Vyatkin, Xiaojing Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

3 Citations (Scopus)

Abstract

Cooling systems consume around 40% of modern data centres' total energy consumption, thus reducing the energy waste in this sector will have positive environmental impact. There exist several cooling methods appropriate for particular conditions. Since environmental conditions, such as air temperature, change on seasonal bases, no single cooling method can be claimed to be the best. In contrast, data centres with multiple cooling methods can perform more efficiently. However, the decision-making process of which cooling system is the most appropriate one should be an automated process. In this paper, a decision-making process based on simulation is proposed. The simulation tool comprises of a mathematical model and a multi-agent control. The mathematical model simulates the thermal behaviour of SICS ICE data centre, which is a real facility located in Northern Sweden. The main aim of the simulation is to calculate thermal conditions and energy consumption of different cooling methods in cold and hot seasons. The result then will be used by multi-agent control to choose the most appropriate cooling method.

Original languageEnglish
Title of host publicationProceedings IECON 2017 - 43rd Annual Conference of the IEEE Industrial Electronics Society
PublisherIEEE
Pages8785-8790
Number of pages6
Volume2017-January
ISBN (Electronic)9781538611272
DOIs
Publication statusPublished - 15 Dec 2017
MoE publication typeA4 Article in a conference publication
EventAnnual Conference of the IEEE Industrial Electronics Society - Beijing, China
Duration: 29 Oct 20171 Nov 2017
Conference number: 43
http://iecon2017.csp.escience.cn/

Publication series

Name Proceedings of the Annual Conference of the IEEE Industrial Electronics Society
PublisherIEEE
ISSN (Print)1553-572X

Conference

ConferenceAnnual Conference of the IEEE Industrial Electronics Society
Abbreviated titleIECON
CountryChina
CityBeijing
Period29/10/201701/11/2017
Internet address

Keywords

  • data center
  • decision making
  • energy efficiency
  • mathematical modelling
  • multi-agent control
  • multiple cooling systems
  • simulation

Fingerprint

Dive into the research topics of 'Energy efficient decision making in data centers with multiple cooling methods'. Together they form a unique fingerprint.

Cite this