Program counter-based prediction techniques for dynamic power management

Christopher Gniady, Ali R. Butt, Y. Charlie Hu, Yung Hsiang Lu

Research output: Contribution to journalArticle

26 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. It 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 higher energy savings while incurring lower mispredictions. We perform an extensive evaluation study of PCAP using a detailed trace-driven simulation and an actual Linux implementation. Our results show that PCAP achieves lower average mispredictions and higher energy savings than the simple timeout scheme and the state-of-the-art Learning Tree scheme.

Original languageEnglish (US)
Pages (from-to)641-658
Number of pages18
JournalIEEE Transactions on Computers
Volume55
Issue number6
DOIs
StatePublished - Jun 2006

Fingerprint

Power Management
Predictors
Prediction
Energy conservation
Energy Saving
Predict
Energy utilization
High Energy
Linux
Power management
Operating Systems
Correlate
Energy Consumption
Trace
Path
Computing
Evaluation
Energy

Keywords

  • Energy-aware systems
  • Hardware/software interfaces
  • Storage management

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Hardware and Architecture

Cite this

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

In: IEEE Transactions on Computers, Vol. 55, No. 6, 06.2006, p. 641-658.

Research output: Contribution to journalArticle

Gniady, Christopher ; Butt, Ali R. ; Hu, Y. Charlie ; Lu, Yung Hsiang. / Program counter-based prediction techniques for dynamic power management. In: IEEE Transactions on Computers. 2006 ; Vol. 55, No. 6. pp. 641-658.
@article{cfd77efabb5a4762838b8c25b94dd723,
title = "Program counter-based prediction 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. It 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 higher energy savings while incurring lower mispredictions. We perform an extensive evaluation study of PCAP using a detailed trace-driven simulation and an actual Linux implementation. Our results show that PCAP achieves lower average mispredictions and higher energy savings than the simple timeout scheme and the state-of-the-art Learning Tree scheme.",
keywords = "Energy-aware systems, Hardware/software interfaces, Storage management",
author = "Christopher Gniady and Butt, {Ali R.} and Hu, {Y. Charlie} and Lu, {Yung Hsiang}",
year = "2006",
month = "6",
doi = "10.1109/TC.2006.87",
language = "English (US)",
volume = "55",
pages = "641--658",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "6",

}

TY - JOUR

T1 - Program counter-based prediction techniques for dynamic power management

AU - Gniady, Christopher

AU - Butt, Ali R.

AU - Hu, Y. Charlie

AU - Lu, Yung Hsiang

PY - 2006/6

Y1 - 2006/6

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. It 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 higher energy savings while incurring lower mispredictions. We perform an extensive evaluation study of PCAP using a detailed trace-driven simulation and an actual Linux implementation. Our results show that PCAP achieves lower average mispredictions and higher energy savings than the simple timeout scheme and the state-of-the-art Learning Tree scheme.

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. It 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 higher energy savings while incurring lower mispredictions. We perform an extensive evaluation study of PCAP using a detailed trace-driven simulation and an actual Linux implementation. Our results show that PCAP achieves lower average mispredictions and higher energy savings than the simple timeout scheme and the state-of-the-art Learning Tree scheme.

KW - Energy-aware systems

KW - Hardware/software interfaces

KW - Storage management

UR - http://www.scopus.com/inward/record.url?scp=33646477108&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646477108&partnerID=8YFLogxK

U2 - 10.1109/TC.2006.87

DO - 10.1109/TC.2006.87

M3 - Article

VL - 55

SP - 641

EP - 658

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 6

ER -