Program counter based techniques for dynamic power management

Christopher Gniady, Y. Charlie Hu, Yung Hsiang Lu

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

30 Citations (Scopus)

Abstract

Reducing energy consumption has become one of the major challenges in designing future computing systems. This paper proposes a novel idea of using program counters to predict I/O activities in the operating system. The paper presents a complete design of Program-Counter Access Predictor (PCAP) that dynamically learns the access patterns of applications and predicts when an I/O device can be shut down to save energy. PCAP uses path-based correlation to observe a particular sequence of program counters leading to each idle period, and predicts future occurrences of that idle period. PCAP differs from previously proposed shutdown predictors in its ability to: (1) correlate I/O operations to particular behavior of the applications and users, (2) carry prediction information across multiple executions of the applications, and (3) attain better energy savings while incurring low mispredictions.

Original languageEnglish (US)
Title of host publicationIEEE High-Performance Computer Architecture Symposium Proceedings
Pages24-35
Number of pages12
Volume10
StatePublished - 2004
Externally publishedYes
EventProceedings - 10th International Symposium on High Performance Computer Architecture - Madrid, Spain
Duration: Feb 14 2004Feb 18 2004

Other

OtherProceedings - 10th International Symposium on High Performance Computer Architecture
CountrySpain
CityMadrid
Period2/14/042/18/04

Fingerprint

Energy conservation
Energy utilization
Power management

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

Gniady, C., Hu, Y. C., & Lu, Y. H. (2004). Program counter based techniques for dynamic power management. In IEEE High-Performance Computer Architecture Symposium Proceedings (Vol. 10, pp. 24-35)

Program counter based techniques for dynamic power management. / Gniady, Christopher; Hu, Y. Charlie; Lu, Yung Hsiang.

IEEE High-Performance Computer Architecture Symposium Proceedings. Vol. 10 2004. p. 24-35.

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

Gniady, C, Hu, YC & Lu, YH 2004, Program counter based techniques for dynamic power management. in IEEE High-Performance Computer Architecture Symposium Proceedings. vol. 10, pp. 24-35, Proceedings - 10th International Symposium on High Performance Computer Architecture, Madrid, Spain, 2/14/04.
Gniady C, Hu YC, Lu YH. Program counter based techniques for dynamic power management. In IEEE High-Performance Computer Architecture Symposium Proceedings. Vol. 10. 2004. p. 24-35
Gniady, Christopher ; Hu, Y. Charlie ; Lu, Yung Hsiang. / Program counter based techniques for dynamic power management. IEEE High-Performance Computer Architecture Symposium Proceedings. Vol. 10 2004. pp. 24-35
@inproceedings{8d61a47580764448b79e3f33f242a6b1,
title = "Program counter based techniques for dynamic power management",
abstract = "Reducing energy consumption has become one of the major challenges in designing future computing systems. This paper proposes a novel idea of using program counters to predict I/O activities in the operating system. The paper presents a complete design of Program-Counter Access Predictor (PCAP) that dynamically learns the access patterns of applications and predicts when an I/O device can be shut down to save energy. PCAP uses path-based correlation to observe a particular sequence of program counters leading to each idle period, and predicts future occurrences of that idle period. PCAP differs from previously proposed shutdown predictors in its ability to: (1) correlate I/O operations to particular behavior of the applications and users, (2) carry prediction information across multiple executions of the applications, and (3) attain better energy savings while incurring low mispredictions.",
author = "Christopher Gniady and Hu, {Y. Charlie} and Lu, {Yung Hsiang}",
year = "2004",
language = "English (US)",
volume = "10",
pages = "24--35",
booktitle = "IEEE High-Performance Computer Architecture Symposium Proceedings",

}

TY - GEN

T1 - Program counter based techniques for dynamic power management

AU - Gniady, Christopher

AU - Hu, Y. Charlie

AU - Lu, Yung Hsiang

PY - 2004

Y1 - 2004

N2 - Reducing energy consumption has become one of the major challenges in designing future computing systems. This paper proposes a novel idea of using program counters to predict I/O activities in the operating system. The paper presents a complete design of Program-Counter Access Predictor (PCAP) that dynamically learns the access patterns of applications and predicts when an I/O device can be shut down to save energy. PCAP uses path-based correlation to observe a particular sequence of program counters leading to each idle period, and predicts future occurrences of that idle period. PCAP differs from previously proposed shutdown predictors in its ability to: (1) correlate I/O operations to particular behavior of the applications and users, (2) carry prediction information across multiple executions of the applications, and (3) attain better energy savings while incurring low mispredictions.

AB - Reducing energy consumption has become one of the major challenges in designing future computing systems. This paper proposes a novel idea of using program counters to predict I/O activities in the operating system. The paper presents a complete design of Program-Counter Access Predictor (PCAP) that dynamically learns the access patterns of applications and predicts when an I/O device can be shut down to save energy. PCAP uses path-based correlation to observe a particular sequence of program counters leading to each idle period, and predicts future occurrences of that idle period. PCAP differs from previously proposed shutdown predictors in its ability to: (1) correlate I/O operations to particular behavior of the applications and users, (2) carry prediction information across multiple executions of the applications, and (3) attain better energy savings while incurring low mispredictions.

UR - http://www.scopus.com/inward/record.url?scp=2342488673&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2342488673&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:2342488673

VL - 10

SP - 24

EP - 35

BT - IEEE High-Performance Computer Architecture Symposium Proceedings

ER -