Data Center power cost optimization via workload modulation

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

5 Citations (Scopus)

Abstract

We formulate optimization problems to study how data centers might modulate their power demands for cost-effective operation taking into account various complexities exhibited by real-world electricity pricing schemes. For computational tractability reasons, we work with a fluid model for power demands which we imagine can be modulated using two abstract knobs of demand dropping and demand delaying (each with its associated penalties or costs). We consider both stochastically known and completely unknown inputs, which are likely to capture different data center scenarios. Using empirical evaluation with both real-world and synthetic power demands and real-world prices, we demonstrate the efficacy of our techniques.

Original languageEnglish (US)
Title of host publicationProceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013
PublisherIEEE Computer Society
Pages260-263
Number of pages4
ISBN (Print)9780769551524
DOIs
StatePublished - Jan 1 2013
Event2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013 - Dresden, Saxony, Germany
Duration: Dec 9 2013Dec 12 2013

Publication series

NameProceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013

Other

Other2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013
CountryGermany
CityDresden, Saxony
Period12/9/1312/12/13

Fingerprint

Modulation
Knobs
Costs
Data acquisition
Electricity
Fluids

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications

Cite this

Wang, C., Urgaonkar, B., Wang, Q., Kesidis, G., & Sivasubramaniam, A. (2013). Data Center power cost optimization via workload modulation. In Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013 (pp. 260-263). [6809409] (Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013). IEEE Computer Society. https://doi.org/10.1109/UCC.2013.52
Wang, Cheng ; Urgaonkar, Bhuvan ; Wang, Qian ; Kesidis, George ; Sivasubramaniam, Anand. / Data Center power cost optimization via workload modulation. Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013. IEEE Computer Society, 2013. pp. 260-263 (Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013).
@inproceedings{92bf3869ab7a4acfad805c4b8ff31b24,
title = "Data Center power cost optimization via workload modulation",
abstract = "We formulate optimization problems to study how data centers might modulate their power demands for cost-effective operation taking into account various complexities exhibited by real-world electricity pricing schemes. For computational tractability reasons, we work with a fluid model for power demands which we imagine can be modulated using two abstract knobs of demand dropping and demand delaying (each with its associated penalties or costs). We consider both stochastically known and completely unknown inputs, which are likely to capture different data center scenarios. Using empirical evaluation with both real-world and synthetic power demands and real-world prices, we demonstrate the efficacy of our techniques.",
author = "Cheng Wang and Bhuvan Urgaonkar and Qian Wang and George Kesidis and Anand Sivasubramaniam",
year = "2013",
month = "1",
day = "1",
doi = "10.1109/UCC.2013.52",
language = "English (US)",
isbn = "9780769551524",
series = "Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013",
publisher = "IEEE Computer Society",
pages = "260--263",
booktitle = "Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013",
address = "United States",

}

Wang, C, Urgaonkar, B, Wang, Q, Kesidis, G & Sivasubramaniam, A 2013, Data Center power cost optimization via workload modulation. in Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013., 6809409, Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013, IEEE Computer Society, pp. 260-263, 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013, Dresden, Saxony, Germany, 12/9/13. https://doi.org/10.1109/UCC.2013.52

Data Center power cost optimization via workload modulation. / Wang, Cheng; Urgaonkar, Bhuvan; Wang, Qian; Kesidis, George; Sivasubramaniam, Anand.

Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013. IEEE Computer Society, 2013. p. 260-263 6809409 (Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013).

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

TY - GEN

T1 - Data Center power cost optimization via workload modulation

AU - Wang, Cheng

AU - Urgaonkar, Bhuvan

AU - Wang, Qian

AU - Kesidis, George

AU - Sivasubramaniam, Anand

PY - 2013/1/1

Y1 - 2013/1/1

N2 - We formulate optimization problems to study how data centers might modulate their power demands for cost-effective operation taking into account various complexities exhibited by real-world electricity pricing schemes. For computational tractability reasons, we work with a fluid model for power demands which we imagine can be modulated using two abstract knobs of demand dropping and demand delaying (each with its associated penalties or costs). We consider both stochastically known and completely unknown inputs, which are likely to capture different data center scenarios. Using empirical evaluation with both real-world and synthetic power demands and real-world prices, we demonstrate the efficacy of our techniques.

AB - We formulate optimization problems to study how data centers might modulate their power demands for cost-effective operation taking into account various complexities exhibited by real-world electricity pricing schemes. For computational tractability reasons, we work with a fluid model for power demands which we imagine can be modulated using two abstract knobs of demand dropping and demand delaying (each with its associated penalties or costs). We consider both stochastically known and completely unknown inputs, which are likely to capture different data center scenarios. Using empirical evaluation with both real-world and synthetic power demands and real-world prices, we demonstrate the efficacy of our techniques.

UR - http://www.scopus.com/inward/record.url?scp=84901652233&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901652233&partnerID=8YFLogxK

U2 - 10.1109/UCC.2013.52

DO - 10.1109/UCC.2013.52

M3 - Conference contribution

SN - 9780769551524

T3 - Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013

SP - 260

EP - 263

BT - Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013

PB - IEEE Computer Society

ER -

Wang C, Urgaonkar B, Wang Q, Kesidis G, Sivasubramaniam A. Data Center power cost optimization via workload modulation. In Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013. IEEE Computer Society. 2013. p. 260-263. 6809409. (Proceedings - 2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013). https://doi.org/10.1109/UCC.2013.52