Predictive temperature-aware DVFS

Jong Sung Lee, Kevin Skadron, Sung Woo Jung

Research output: Contribution to journalArticle

52 Citations (Scopus)

Abstract

In this paper, we propose predictive temperature-aware Dynamic Voltage and Frequency Scaling (DVFS) using the performance counters that are already embedded in commercial microprocessors. By using the performance counters and simple regression analysis, we can predict the localized temperature and efficiently scale the voltage/frequency. When localized thermal problems that were not detected by thermal sensors are found after layout (or fabrication), the thermal problems can be avoided by the proposed software solution without delaying time-to-market. The evaluation results show that in a Linux-based laptop with the Intel Core2 Duo processor, DVFS using the performance counters performs comparable to DVFS using the thermal sensor.

Original languageEnglish
Article number5255230
Pages (from-to)127-133
Number of pages7
JournalIEEE Transactions on Computers
Volume59
Issue number1
DOIs
Publication statusPublished - 2010 Jan 4

Fingerprint

Heat problems
Voltage
Scaling
Sensors
Regression analysis
Sensor
Temperature
Microprocessor chips
Microprocessor
Linux
Regression Analysis
Layout
Fabrication
Electric potential
Predict
Software
Voltage scaling
Dynamic frequency scaling
Evaluation
Hot Temperature

Keywords

  • DVFS
  • Dynamic thermal management
  • Microprocessor
  • Performance counter
  • Thermal sensor

ASJC Scopus subject areas

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

Cite this

Predictive temperature-aware DVFS. / Lee, Jong Sung; Skadron, Kevin; Jung, Sung Woo.

In: IEEE Transactions on Computers, Vol. 59, No. 1, 5255230, 04.01.2010, p. 127-133.

Research output: Contribution to journalArticle

Lee, Jong Sung ; Skadron, Kevin ; Jung, Sung Woo. / Predictive temperature-aware DVFS. In: IEEE Transactions on Computers. 2010 ; Vol. 59, No. 1. pp. 127-133.
@article{90a4ab1b7ac843f9ae3a98fba62d7bb6,
title = "Predictive temperature-aware DVFS",
abstract = "In this paper, we propose predictive temperature-aware Dynamic Voltage and Frequency Scaling (DVFS) using the performance counters that are already embedded in commercial microprocessors. By using the performance counters and simple regression analysis, we can predict the localized temperature and efficiently scale the voltage/frequency. When localized thermal problems that were not detected by thermal sensors are found after layout (or fabrication), the thermal problems can be avoided by the proposed software solution without delaying time-to-market. The evaluation results show that in a Linux-based laptop with the Intel Core2 Duo processor, DVFS using the performance counters performs comparable to DVFS using the thermal sensor.",
keywords = "DVFS, Dynamic thermal management, Microprocessor, Performance counter, Thermal sensor",
author = "Lee, {Jong Sung} and Kevin Skadron and Jung, {Sung Woo}",
year = "2010",
month = "1",
day = "4",
doi = "10.1109/TC.2009.136",
language = "English",
volume = "59",
pages = "127--133",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "1",

}

TY - JOUR

T1 - Predictive temperature-aware DVFS

AU - Lee, Jong Sung

AU - Skadron, Kevin

AU - Jung, Sung Woo

PY - 2010/1/4

Y1 - 2010/1/4

N2 - In this paper, we propose predictive temperature-aware Dynamic Voltage and Frequency Scaling (DVFS) using the performance counters that are already embedded in commercial microprocessors. By using the performance counters and simple regression analysis, we can predict the localized temperature and efficiently scale the voltage/frequency. When localized thermal problems that were not detected by thermal sensors are found after layout (or fabrication), the thermal problems can be avoided by the proposed software solution without delaying time-to-market. The evaluation results show that in a Linux-based laptop with the Intel Core2 Duo processor, DVFS using the performance counters performs comparable to DVFS using the thermal sensor.

AB - In this paper, we propose predictive temperature-aware Dynamic Voltage and Frequency Scaling (DVFS) using the performance counters that are already embedded in commercial microprocessors. By using the performance counters and simple regression analysis, we can predict the localized temperature and efficiently scale the voltage/frequency. When localized thermal problems that were not detected by thermal sensors are found after layout (or fabrication), the thermal problems can be avoided by the proposed software solution without delaying time-to-market. The evaluation results show that in a Linux-based laptop with the Intel Core2 Duo processor, DVFS using the performance counters performs comparable to DVFS using the thermal sensor.

KW - DVFS

KW - Dynamic thermal management

KW - Microprocessor

KW - Performance counter

KW - Thermal sensor

UR - http://www.scopus.com/inward/record.url?scp=72949089066&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=72949089066&partnerID=8YFLogxK

U2 - 10.1109/TC.2009.136

DO - 10.1109/TC.2009.136

M3 - Article

VL - 59

SP - 127

EP - 133

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 1

M1 - 5255230

ER -