Function-level power estimation methodology for microprocessors

Gang Qu, Naoyuki Kawabe, Kimiyoshi Usami, Miodrag Potkonjak

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

66 Citations (Scopus)

Abstract

We have developed a function-level power estimation methodology for predicting the power dissipation of embedded software. For a given microprocessor core, we empirically build the 'power data bank', which stores the power information of the built-in library functions and basic instructions. To estimate the average power of an embedded software on this core, we first get the execution information of the target software from program profiling/tracing tools. Then we evaluate the total energy consumption and execution time based on the 'power data bank', and take their ratio as the average power. High efficiency is achieved because no power simulator is used once the 'power data bank' is built. We apply this method to a commercial microprocessor core and get power estimates with an average error of 3%. With this method, microprocessor vendors can provide users the 'power data bank' without releasing details of the core to help users get early power estimates and eventually guide power optimization.

Original languageEnglish
Title of host publicationProceedings - Design Automation Conference
PublisherIEEE
Pages810-813
Number of pages4
Publication statusPublished - 2000
Externally publishedYes
EventDAC 2000: 37th Design Automation Conference - Los Angeles, CA, USA
Duration: 2000 Jun 52000 Jun 9

Other

OtherDAC 2000: 37th Design Automation Conference
CityLos Angeles, CA, USA
Period00/6/500/6/9

Fingerprint

Microprocessor chips
Embedded software
Energy dissipation
Energy utilization
Simulators

ASJC Scopus subject areas

  • Hardware and Architecture
  • Control and Systems Engineering

Cite this

Qu, G., Kawabe, N., Usami, K., & Potkonjak, M. (2000). Function-level power estimation methodology for microprocessors. In Proceedings - Design Automation Conference (pp. 810-813). IEEE.

Function-level power estimation methodology for microprocessors. / Qu, Gang; Kawabe, Naoyuki; Usami, Kimiyoshi; Potkonjak, Miodrag.

Proceedings - Design Automation Conference. IEEE, 2000. p. 810-813.

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

Qu, G, Kawabe, N, Usami, K & Potkonjak, M 2000, Function-level power estimation methodology for microprocessors. in Proceedings - Design Automation Conference. IEEE, pp. 810-813, DAC 2000: 37th Design Automation Conference, Los Angeles, CA, USA, 00/6/5.
Qu G, Kawabe N, Usami K, Potkonjak M. Function-level power estimation methodology for microprocessors. In Proceedings - Design Automation Conference. IEEE. 2000. p. 810-813
Qu, Gang ; Kawabe, Naoyuki ; Usami, Kimiyoshi ; Potkonjak, Miodrag. / Function-level power estimation methodology for microprocessors. Proceedings - Design Automation Conference. IEEE, 2000. pp. 810-813
@inproceedings{8db853b173d749beb08a49aec500998f,
title = "Function-level power estimation methodology for microprocessors",
abstract = "We have developed a function-level power estimation methodology for predicting the power dissipation of embedded software. For a given microprocessor core, we empirically build the 'power data bank', which stores the power information of the built-in library functions and basic instructions. To estimate the average power of an embedded software on this core, we first get the execution information of the target software from program profiling/tracing tools. Then we evaluate the total energy consumption and execution time based on the 'power data bank', and take their ratio as the average power. High efficiency is achieved because no power simulator is used once the 'power data bank' is built. We apply this method to a commercial microprocessor core and get power estimates with an average error of 3{\%}. With this method, microprocessor vendors can provide users the 'power data bank' without releasing details of the core to help users get early power estimates and eventually guide power optimization.",
author = "Gang Qu and Naoyuki Kawabe and Kimiyoshi Usami and Miodrag Potkonjak",
year = "2000",
language = "English",
pages = "810--813",
booktitle = "Proceedings - Design Automation Conference",
publisher = "IEEE",

}

TY - GEN

T1 - Function-level power estimation methodology for microprocessors

AU - Qu, Gang

AU - Kawabe, Naoyuki

AU - Usami, Kimiyoshi

AU - Potkonjak, Miodrag

PY - 2000

Y1 - 2000

N2 - We have developed a function-level power estimation methodology for predicting the power dissipation of embedded software. For a given microprocessor core, we empirically build the 'power data bank', which stores the power information of the built-in library functions and basic instructions. To estimate the average power of an embedded software on this core, we first get the execution information of the target software from program profiling/tracing tools. Then we evaluate the total energy consumption and execution time based on the 'power data bank', and take their ratio as the average power. High efficiency is achieved because no power simulator is used once the 'power data bank' is built. We apply this method to a commercial microprocessor core and get power estimates with an average error of 3%. With this method, microprocessor vendors can provide users the 'power data bank' without releasing details of the core to help users get early power estimates and eventually guide power optimization.

AB - We have developed a function-level power estimation methodology for predicting the power dissipation of embedded software. For a given microprocessor core, we empirically build the 'power data bank', which stores the power information of the built-in library functions and basic instructions. To estimate the average power of an embedded software on this core, we first get the execution information of the target software from program profiling/tracing tools. Then we evaluate the total energy consumption and execution time based on the 'power data bank', and take their ratio as the average power. High efficiency is achieved because no power simulator is used once the 'power data bank' is built. We apply this method to a commercial microprocessor core and get power estimates with an average error of 3%. With this method, microprocessor vendors can provide users the 'power data bank' without releasing details of the core to help users get early power estimates and eventually guide power optimization.

UR - http://www.scopus.com/inward/record.url?scp=0033685331&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033685331&partnerID=8YFLogxK

M3 - Conference contribution

SP - 810

EP - 813

BT - Proceedings - Design Automation Conference

PB - IEEE

ER -