Providing RS Participation for Geo-Distributed Data Centers Using Deep Learning-Based Power Prediction

Somayyeh Taheri, Maziar Goudarzi, Osamu Yoshie

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

Abstract

Nowadays, geo-distributed Data Centers (DCs) are very common, because of providing more energy efficiency, higher system availability as well as flexibility. In a geo-distributed cloud, each local DC responds to the specific portion of the incoming load which distributed based on different Geographically Load Balancing (GLB) policies. As a large yet flexible power consumer, the local DC has a great impact on the local power grid. From this point of view, a local DC is a good candidate to participate in the emerging power market such as Regulation Service (RS) opportunity, that brings monetary benefits both for the DC as well as the grid. However, a fruitful collaboration requires the DC to have the capability of forecasting its future power consumption. While, given the different GLB policies, the amount of delivered load toward each local DC is a function of the whole system’s conditions, rather than the local situation. Thereby, the problem of RS participation for local DCs in a geo-distributed cloud is challenging. Motivated by this fact, this paper benefits from deep learning to predict the local DCs’ power consumption. We consider two main GLB policies, including Power-aware as well as Cost-aware, to acquire training data and construct a prediction model accordingly. Afterward, we leverage the prediction results to provide the opportunity of RS participation for geo-distributed DCs. Results show that the proposed approach reduces the energy cost by 22% on average in compared with well-known GLB policies.

Original languageEnglish
Title of host publicationHigh-Performance Computing and Big Data Analysis- 2nd International Congress, TopHPC 2019, Revised Selected Papers
EditorsLucio Grandinetti, Reza Shahbazian, Seyedeh Leili Mirtaheri
PublisherSpringer
Pages3-17
Number of pages15
ISBN (Print)9783030334949
DOIs
Publication statusPublished - 2019
Event2nd International Congress on High-Performance Computing and Big Data Analysis, TopHPC 2019 - Tehran, Iran, Islamic Republic of
Duration: 2019 Apr 232019 Apr 25

Publication series

NameCommunications in Computer and Information Science
Volume891
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference2nd International Congress on High-Performance Computing and Big Data Analysis, TopHPC 2019
CountryIran, Islamic Republic of
CityTehran
Period19/4/2319/4/25

Keywords

  • Deep learning prediction
  • Emerging power market
  • Geo-distributed data center
  • Regulation service

ASJC Scopus subject areas

  • Computer Science(all)
  • Mathematics(all)

Fingerprint Dive into the research topics of 'Providing RS Participation for Geo-Distributed Data Centers Using Deep Learning-Based Power Prediction'. Together they form a unique fingerprint.

  • Cite this

    Taheri, S., Goudarzi, M., & Yoshie, O. (2019). Providing RS Participation for Geo-Distributed Data Centers Using Deep Learning-Based Power Prediction. In L. Grandinetti, R. Shahbazian, & S. L. Mirtaheri (Eds.), High-Performance Computing and Big Data Analysis- 2nd International Congress, TopHPC 2019, Revised Selected Papers (pp. 3-17). (Communications in Computer and Information Science; Vol. 891). Springer. https://doi.org/10.1007/978-3-030-33495-6_1