Optimal Peak Shaving Using Batteries at Datacenters: Charging Risk and Degradation Model

Neda Nasiriani, George Kesidis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A datacenter's power consumption is a major contributor to its operational expenditures (op-ex) as determined by peak-demand-over-billing-period based pricing which is often employed by major electric utility providers. There is a growing interest in reducing a datacenters electricity costs by using demand-throttling techniques and/or energy storage devices (batteries which are readily available at most datacenters as a backup energy source). For the latter, we present a Markov Decision Process framework based on power-demand uncertainty and a linearized battery degradation model. This framework also explicitly considers risk of over or under charging the battery resulting in higher cost-savings (up to 2×) with tractable risk. We show the complete characterization of risk-cost trade-off and cost-per-risk as a function of datacenter's workload characteristics. We also, study a linearized battery degradation model empirically, and show the accuracy of this model for bursty workloads, however there is some discrepancy between the linearized model and reality for workloads with lower variability.

Original languageEnglish (US)
Title of host publication2018 International Conference on Computing, Networking and Communications, ICNC 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages58-62
Number of pages5
ISBN (Electronic)9781538636527
DOIs
StatePublished - Jun 19 2018
Event2018 International Conference on Computing, Networking and Communications, ICNC 2018 - Maui, United States
Duration: Mar 5 2018Mar 8 2018

Publication series

Name2018 International Conference on Computing, Networking and Communications, ICNC 2018

Other

Other2018 International Conference on Computing, Networking and Communications, ICNC 2018
CountryUnited States
CityMaui
Period3/5/183/8/18

Fingerprint

Degradation
Costs
Electric utilities
Energy storage
Electric power utilization
Electricity

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

Nasiriani, N., & Kesidis, G. (2018). Optimal Peak Shaving Using Batteries at Datacenters: Charging Risk and Degradation Model. In 2018 International Conference on Computing, Networking and Communications, ICNC 2018 (pp. 58-62). (2018 International Conference on Computing, Networking and Communications, ICNC 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCNC.2018.8390416
Nasiriani, Neda ; Kesidis, George. / Optimal Peak Shaving Using Batteries at Datacenters : Charging Risk and Degradation Model. 2018 International Conference on Computing, Networking and Communications, ICNC 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 58-62 (2018 International Conference on Computing, Networking and Communications, ICNC 2018).
@inproceedings{26348c121ba94bf38a2197e2bba4ce28,
title = "Optimal Peak Shaving Using Batteries at Datacenters: Charging Risk and Degradation Model",
abstract = "A datacenter's power consumption is a major contributor to its operational expenditures (op-ex) as determined by peak-demand-over-billing-period based pricing which is often employed by major electric utility providers. There is a growing interest in reducing a datacenters electricity costs by using demand-throttling techniques and/or energy storage devices (batteries which are readily available at most datacenters as a backup energy source). For the latter, we present a Markov Decision Process framework based on power-demand uncertainty and a linearized battery degradation model. This framework also explicitly considers risk of over or under charging the battery resulting in higher cost-savings (up to 2×) with tractable risk. We show the complete characterization of risk-cost trade-off and cost-per-risk as a function of datacenter's workload characteristics. We also, study a linearized battery degradation model empirically, and show the accuracy of this model for bursty workloads, however there is some discrepancy between the linearized model and reality for workloads with lower variability.",
author = "Neda Nasiriani and George Kesidis",
year = "2018",
month = "6",
day = "19",
doi = "10.1109/ICCNC.2018.8390416",
language = "English (US)",
series = "2018 International Conference on Computing, Networking and Communications, ICNC 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "58--62",
booktitle = "2018 International Conference on Computing, Networking and Communications, ICNC 2018",
address = "United States",

}

Nasiriani, N & Kesidis, G 2018, Optimal Peak Shaving Using Batteries at Datacenters: Charging Risk and Degradation Model. in 2018 International Conference on Computing, Networking and Communications, ICNC 2018. 2018 International Conference on Computing, Networking and Communications, ICNC 2018, Institute of Electrical and Electronics Engineers Inc., pp. 58-62, 2018 International Conference on Computing, Networking and Communications, ICNC 2018, Maui, United States, 3/5/18. https://doi.org/10.1109/ICCNC.2018.8390416

Optimal Peak Shaving Using Batteries at Datacenters : Charging Risk and Degradation Model. / Nasiriani, Neda; Kesidis, George.

2018 International Conference on Computing, Networking and Communications, ICNC 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 58-62 (2018 International Conference on Computing, Networking and Communications, ICNC 2018).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Optimal Peak Shaving Using Batteries at Datacenters

T2 - Charging Risk and Degradation Model

AU - Nasiriani, Neda

AU - Kesidis, George

PY - 2018/6/19

Y1 - 2018/6/19

N2 - A datacenter's power consumption is a major contributor to its operational expenditures (op-ex) as determined by peak-demand-over-billing-period based pricing which is often employed by major electric utility providers. There is a growing interest in reducing a datacenters electricity costs by using demand-throttling techniques and/or energy storage devices (batteries which are readily available at most datacenters as a backup energy source). For the latter, we present a Markov Decision Process framework based on power-demand uncertainty and a linearized battery degradation model. This framework also explicitly considers risk of over or under charging the battery resulting in higher cost-savings (up to 2×) with tractable risk. We show the complete characterization of risk-cost trade-off and cost-per-risk as a function of datacenter's workload characteristics. We also, study a linearized battery degradation model empirically, and show the accuracy of this model for bursty workloads, however there is some discrepancy between the linearized model and reality for workloads with lower variability.

AB - A datacenter's power consumption is a major contributor to its operational expenditures (op-ex) as determined by peak-demand-over-billing-period based pricing which is often employed by major electric utility providers. There is a growing interest in reducing a datacenters electricity costs by using demand-throttling techniques and/or energy storage devices (batteries which are readily available at most datacenters as a backup energy source). For the latter, we present a Markov Decision Process framework based on power-demand uncertainty and a linearized battery degradation model. This framework also explicitly considers risk of over or under charging the battery resulting in higher cost-savings (up to 2×) with tractable risk. We show the complete characterization of risk-cost trade-off and cost-per-risk as a function of datacenter's workload characteristics. We also, study a linearized battery degradation model empirically, and show the accuracy of this model for bursty workloads, however there is some discrepancy between the linearized model and reality for workloads with lower variability.

UR - http://www.scopus.com/inward/record.url?scp=85050121560&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050121560&partnerID=8YFLogxK

U2 - 10.1109/ICCNC.2018.8390416

DO - 10.1109/ICCNC.2018.8390416

M3 - Conference contribution

AN - SCOPUS:85050121560

T3 - 2018 International Conference on Computing, Networking and Communications, ICNC 2018

SP - 58

EP - 62

BT - 2018 International Conference on Computing, Networking and Communications, ICNC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Nasiriani N, Kesidis G. Optimal Peak Shaving Using Batteries at Datacenters: Charging Risk and Degradation Model. In 2018 International Conference on Computing, Networking and Communications, ICNC 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 58-62. (2018 International Conference on Computing, Networking and Communications, ICNC 2018). https://doi.org/10.1109/ICCNC.2018.8390416