Power consumption prediction and power-aware packing in consolidated environments

Jeonghwan Choi, Sriram Govindan, Jinkyu Jeong, Bhuvan Urgaonkar, Anand Sivasubramaniam

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

Consolidation of workloads has emerged as a key mechanism to dampen the rapidly growing energy expenditure within enterprise-scale data centers. To gainfully utilize consolidation-based techniques, we must be able to characterize the power consumption of groups of colocated applications. Such characterization is crucial for effective prediction and enforcement of appropriate limits on power consumption-power budgets-within the data center. We identify two kinds of power budgets: 1) an average budget to capture an upper bound on long-term energy consumption within that level and 2) a sustained budget to capture any restrictions on sustained draw of current above a certain threshold. Using a simple measurement infrastructure, we derive power profile-sstatistical descriptions of the power consumption of applications. Based on insights gained from detailed profiling of several applicationsboth individual and consolidated-we develop models for predicting average and sustained power consumption of consolidated applications. We conduct an experimental evaluation of our techniques on a Xen-based server that consolidates applications drawn from a diverse pool. For a variety of consolidation scenarios, we are able to predict average power consumption within five percent error margin and sustained power within 10 percent error margin. Using prediction techniques allows us to ensure safe yet efficient system operation-in a representative case, we are able to improve the number of applications consolidated on a server from two to three (compared to existing baseline techniques) by choosing the appropriate power state that satisfies the power budgets associated with the server.

Original languageEnglish (US)
Article number5453350
Pages (from-to)1640-1654
Number of pages15
JournalIEEE Transactions on Computers
Volume59
Issue number12
DOIs
StatePublished - Nov 12 2010

Fingerprint

Packing
Power Consumption
Electric power utilization
Consolidation
Prediction
Servers
Server
Data Center
Margin
Percent
Profiling
Experimental Evaluation
Energy Consumption
Workload
Baseline
Energy utilization
Infrastructure
Upper bound
Restriction
Predict

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computational Theory and Mathematics

Cite this

@article{e650caac2a6244b8981bbaa18dc0acb7,
title = "Power consumption prediction and power-aware packing in consolidated environments",
abstract = "Consolidation of workloads has emerged as a key mechanism to dampen the rapidly growing energy expenditure within enterprise-scale data centers. To gainfully utilize consolidation-based techniques, we must be able to characterize the power consumption of groups of colocated applications. Such characterization is crucial for effective prediction and enforcement of appropriate limits on power consumption-power budgets-within the data center. We identify two kinds of power budgets: 1) an average budget to capture an upper bound on long-term energy consumption within that level and 2) a sustained budget to capture any restrictions on sustained draw of current above a certain threshold. Using a simple measurement infrastructure, we derive power profile-sstatistical descriptions of the power consumption of applications. Based on insights gained from detailed profiling of several applicationsboth individual and consolidated-we develop models for predicting average and sustained power consumption of consolidated applications. We conduct an experimental evaluation of our techniques on a Xen-based server that consolidates applications drawn from a diverse pool. For a variety of consolidation scenarios, we are able to predict average power consumption within five percent error margin and sustained power within 10 percent error margin. Using prediction techniques allows us to ensure safe yet efficient system operation-in a representative case, we are able to improve the number of applications consolidated on a server from two to three (compared to existing baseline techniques) by choosing the appropriate power state that satisfies the power budgets associated with the server.",
author = "Jeonghwan Choi and Sriram Govindan and Jinkyu Jeong and Bhuvan Urgaonkar and Anand Sivasubramaniam",
year = "2010",
month = "11",
day = "12",
doi = "10.1109/TC.2010.91",
language = "English (US)",
volume = "59",
pages = "1640--1654",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "12",

}

Power consumption prediction and power-aware packing in consolidated environments. / Choi, Jeonghwan; Govindan, Sriram; Jeong, Jinkyu; Urgaonkar, Bhuvan; Sivasubramaniam, Anand.

In: IEEE Transactions on Computers, Vol. 59, No. 12, 5453350, 12.11.2010, p. 1640-1654.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Power consumption prediction and power-aware packing in consolidated environments

AU - Choi, Jeonghwan

AU - Govindan, Sriram

AU - Jeong, Jinkyu

AU - Urgaonkar, Bhuvan

AU - Sivasubramaniam, Anand

PY - 2010/11/12

Y1 - 2010/11/12

N2 - Consolidation of workloads has emerged as a key mechanism to dampen the rapidly growing energy expenditure within enterprise-scale data centers. To gainfully utilize consolidation-based techniques, we must be able to characterize the power consumption of groups of colocated applications. Such characterization is crucial for effective prediction and enforcement of appropriate limits on power consumption-power budgets-within the data center. We identify two kinds of power budgets: 1) an average budget to capture an upper bound on long-term energy consumption within that level and 2) a sustained budget to capture any restrictions on sustained draw of current above a certain threshold. Using a simple measurement infrastructure, we derive power profile-sstatistical descriptions of the power consumption of applications. Based on insights gained from detailed profiling of several applicationsboth individual and consolidated-we develop models for predicting average and sustained power consumption of consolidated applications. We conduct an experimental evaluation of our techniques on a Xen-based server that consolidates applications drawn from a diverse pool. For a variety of consolidation scenarios, we are able to predict average power consumption within five percent error margin and sustained power within 10 percent error margin. Using prediction techniques allows us to ensure safe yet efficient system operation-in a representative case, we are able to improve the number of applications consolidated on a server from two to three (compared to existing baseline techniques) by choosing the appropriate power state that satisfies the power budgets associated with the server.

AB - Consolidation of workloads has emerged as a key mechanism to dampen the rapidly growing energy expenditure within enterprise-scale data centers. To gainfully utilize consolidation-based techniques, we must be able to characterize the power consumption of groups of colocated applications. Such characterization is crucial for effective prediction and enforcement of appropriate limits on power consumption-power budgets-within the data center. We identify two kinds of power budgets: 1) an average budget to capture an upper bound on long-term energy consumption within that level and 2) a sustained budget to capture any restrictions on sustained draw of current above a certain threshold. Using a simple measurement infrastructure, we derive power profile-sstatistical descriptions of the power consumption of applications. Based on insights gained from detailed profiling of several applicationsboth individual and consolidated-we develop models for predicting average and sustained power consumption of consolidated applications. We conduct an experimental evaluation of our techniques on a Xen-based server that consolidates applications drawn from a diverse pool. For a variety of consolidation scenarios, we are able to predict average power consumption within five percent error margin and sustained power within 10 percent error margin. Using prediction techniques allows us to ensure safe yet efficient system operation-in a representative case, we are able to improve the number of applications consolidated on a server from two to three (compared to existing baseline techniques) by choosing the appropriate power state that satisfies the power budgets associated with the server.

UR - http://www.scopus.com/inward/record.url?scp=78149264795&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78149264795&partnerID=8YFLogxK

U2 - 10.1109/TC.2010.91

DO - 10.1109/TC.2010.91

M3 - Article

AN - SCOPUS:78149264795

VL - 59

SP - 1640

EP - 1654

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 12

M1 - 5453350

ER -