Impact of workload and renewable prediction on the value of geographical workload management

Zahra Abbasi, Madhurima Pore, Sandeep Gupta

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

3 Citations (Scopus)

Abstract

There has been increasing demand for energy sustainable and low-cost operation in cloud computing. This paper proposes dynamic Geographical Load Balancing and energy buffering management (GLB) to achieve these goals which (i) shifts workload (particularly peak workload demand) toward Data centers that offer low utility rate or green energy at a time, and (ii) banks excess green and low-cost energy to shift peak workload demand away from high utility rate. Such a scheme needs to be aware of the workload intensity and the available renewable power of the cloud in future (over a relatively long prediction window such as a day). Existing solutions mainly focus on developing algorithms and demonstrating the cost efficiency of GLB, disregarding the prediction accuracy of the workload and the renewable power. However, erroneous information decreases the efficiency of GLB. This paper studies the performance of the online GLB solution when using time-series based prediction techniques (e.g., ARIMA ) for the workload and the renewable power (i.e., solar and wind). The results of the simulation study using realistic traces highlight that GLB with and without prediction error is effective in reducing average energy cost and increasing sustainability of data centers. Further, GLB is shown to be significantly effective in shaving peak power draw from the grid (e.g., reducing peak power upto 100%), however the erroneous information due to the prediction error adversely affects its performance. Furthermore, the simulation study indicates that the optimal mix of the renewable power (i.e., wind and solar) to be leveraged by GLB, is achieved when data centers are powered from both the solar and the wind power.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages1-15
Number of pages15
Volume8343 LNCS
ISBN (Print)9783642551482
DOIs
StatePublished - 2014
Event2nd International Workshop on Energy-Efficient Data Centers, E2DC 2013 - Berkeley, CA, United States
Duration: May 21 2013May 21 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8343 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other2nd International Workshop on Energy-Efficient Data Centers, E2DC 2013
CountryUnited States
CityBerkeley, CA
Period5/21/135/21/13

Fingerprint

Workload
Prediction
Data Center
Wind power
Wind Power
Solar energy
Prediction Error
Costs
Energy
Simulation Study
ARIMA
Energy management
Cost Efficiency
Cloud computing
Energy Management
Sustainability
Resource allocation
Sustainable development
Time series
Cloud Computing

Keywords

  • Cloud computing
  • Data Centers
  • electricity cost
  • Energy Management
  • Energy Storage
  • Renewable power
  • Workload Prediction

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Abbasi, Z., Pore, M., & Gupta, S. (2014). Impact of workload and renewable prediction on the value of geographical workload management. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8343 LNCS, pp. 1-15). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8343 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-642-55149-9_1

Impact of workload and renewable prediction on the value of geographical workload management. / Abbasi, Zahra; Pore, Madhurima; Gupta, Sandeep.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8343 LNCS Springer Verlag, 2014. p. 1-15 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8343 LNCS).

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

Abbasi, Z, Pore, M & Gupta, S 2014, Impact of workload and renewable prediction on the value of geographical workload management. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 8343 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8343 LNCS, Springer Verlag, pp. 1-15, 2nd International Workshop on Energy-Efficient Data Centers, E2DC 2013, Berkeley, CA, United States, 5/21/13. https://doi.org/10.1007/978-3-642-55149-9_1
Abbasi Z, Pore M, Gupta S. Impact of workload and renewable prediction on the value of geographical workload management. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8343 LNCS. Springer Verlag. 2014. p. 1-15. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-55149-9_1
Abbasi, Zahra ; Pore, Madhurima ; Gupta, Sandeep. / Impact of workload and renewable prediction on the value of geographical workload management. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8343 LNCS Springer Verlag, 2014. pp. 1-15 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{9229e017895d47709edb74c15d43dd10,
title = "Impact of workload and renewable prediction on the value of geographical workload management",
abstract = "There has been increasing demand for energy sustainable and low-cost operation in cloud computing. This paper proposes dynamic Geographical Load Balancing and energy buffering management (GLB) to achieve these goals which (i) shifts workload (particularly peak workload demand) toward Data centers that offer low utility rate or green energy at a time, and (ii) banks excess green and low-cost energy to shift peak workload demand away from high utility rate. Such a scheme needs to be aware of the workload intensity and the available renewable power of the cloud in future (over a relatively long prediction window such as a day). Existing solutions mainly focus on developing algorithms and demonstrating the cost efficiency of GLB, disregarding the prediction accuracy of the workload and the renewable power. However, erroneous information decreases the efficiency of GLB. This paper studies the performance of the online GLB solution when using time-series based prediction techniques (e.g., ARIMA ) for the workload and the renewable power (i.e., solar and wind). The results of the simulation study using realistic traces highlight that GLB with and without prediction error is effective in reducing average energy cost and increasing sustainability of data centers. Further, GLB is shown to be significantly effective in shaving peak power draw from the grid (e.g., reducing peak power upto 100{\%}), however the erroneous information due to the prediction error adversely affects its performance. Furthermore, the simulation study indicates that the optimal mix of the renewable power (i.e., wind and solar) to be leveraged by GLB, is achieved when data centers are powered from both the solar and the wind power.",
keywords = "Cloud computing, Data Centers, electricity cost, Energy Management, Energy Storage, Renewable power, Workload Prediction",
author = "Zahra Abbasi and Madhurima Pore and Sandeep Gupta",
year = "2014",
doi = "10.1007/978-3-642-55149-9_1",
language = "English (US)",
isbn = "9783642551482",
volume = "8343 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "1--15",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Impact of workload and renewable prediction on the value of geographical workload management

AU - Abbasi, Zahra

AU - Pore, Madhurima

AU - Gupta, Sandeep

PY - 2014

Y1 - 2014

N2 - There has been increasing demand for energy sustainable and low-cost operation in cloud computing. This paper proposes dynamic Geographical Load Balancing and energy buffering management (GLB) to achieve these goals which (i) shifts workload (particularly peak workload demand) toward Data centers that offer low utility rate or green energy at a time, and (ii) banks excess green and low-cost energy to shift peak workload demand away from high utility rate. Such a scheme needs to be aware of the workload intensity and the available renewable power of the cloud in future (over a relatively long prediction window such as a day). Existing solutions mainly focus on developing algorithms and demonstrating the cost efficiency of GLB, disregarding the prediction accuracy of the workload and the renewable power. However, erroneous information decreases the efficiency of GLB. This paper studies the performance of the online GLB solution when using time-series based prediction techniques (e.g., ARIMA ) for the workload and the renewable power (i.e., solar and wind). The results of the simulation study using realistic traces highlight that GLB with and without prediction error is effective in reducing average energy cost and increasing sustainability of data centers. Further, GLB is shown to be significantly effective in shaving peak power draw from the grid (e.g., reducing peak power upto 100%), however the erroneous information due to the prediction error adversely affects its performance. Furthermore, the simulation study indicates that the optimal mix of the renewable power (i.e., wind and solar) to be leveraged by GLB, is achieved when data centers are powered from both the solar and the wind power.

AB - There has been increasing demand for energy sustainable and low-cost operation in cloud computing. This paper proposes dynamic Geographical Load Balancing and energy buffering management (GLB) to achieve these goals which (i) shifts workload (particularly peak workload demand) toward Data centers that offer low utility rate or green energy at a time, and (ii) banks excess green and low-cost energy to shift peak workload demand away from high utility rate. Such a scheme needs to be aware of the workload intensity and the available renewable power of the cloud in future (over a relatively long prediction window such as a day). Existing solutions mainly focus on developing algorithms and demonstrating the cost efficiency of GLB, disregarding the prediction accuracy of the workload and the renewable power. However, erroneous information decreases the efficiency of GLB. This paper studies the performance of the online GLB solution when using time-series based prediction techniques (e.g., ARIMA ) for the workload and the renewable power (i.e., solar and wind). The results of the simulation study using realistic traces highlight that GLB with and without prediction error is effective in reducing average energy cost and increasing sustainability of data centers. Further, GLB is shown to be significantly effective in shaving peak power draw from the grid (e.g., reducing peak power upto 100%), however the erroneous information due to the prediction error adversely affects its performance. Furthermore, the simulation study indicates that the optimal mix of the renewable power (i.e., wind and solar) to be leveraged by GLB, is achieved when data centers are powered from both the solar and the wind power.

KW - Cloud computing

KW - Data Centers

KW - electricity cost

KW - Energy Management

KW - Energy Storage

KW - Renewable power

KW - Workload Prediction

UR - http://www.scopus.com/inward/record.url?scp=84902597746&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84902597746&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-55149-9_1

DO - 10.1007/978-3-642-55149-9_1

M3 - Conference contribution

AN - SCOPUS:84902597746

SN - 9783642551482

VL - 8343 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 1

EP - 15

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -