Automatic source code specialization for energy reduction

E. Y. Chung, L. Benini, G. De Micheli

Research output: Contribution to conferencePaper

7 Citations (Scopus)

Abstract

This paper presents a framework to reduce the computational effort of software programs, using value profiling and partial evaluation. Our tool reduces computational effort by specializing a program for highly expected situations and such a reduction translates into both energy and performance improvement. Procedure calls executed frequently with same parameter values are defined as highly expected situations (common cases). The choice of the best transformation of common cases is achieved by solving three search problems. The first identifies effective common cases to be specialized, the second searches for an optimal solution for effective common case, and the third examines the interplay among the specialized cases. Our technique improves both energy consumption and performance of the source code up to more than twice and in average about 25% over the original program. Also, our pruning techniques reduce the searching time by 80% compared to exhaustive approach.

Original languageEnglish
Pages80-83
Number of pages4
Publication statusPublished - 2001 Jan 1
EventInternational Symposium on Low Electronics and Design (ISLPED'01) - Huntington Beach, CA, United States
Duration: 2001 Aug 62001 Aug 7

Other

OtherInternational Symposium on Low Electronics and Design (ISLPED'01)
CountryUnited States
CityHuntington Beach, CA
Period01/8/601/8/7

Fingerprint

Energy utilization

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Chung, E. Y., Benini, L., & De Micheli, G. (2001). Automatic source code specialization for energy reduction. 80-83. Paper presented at International Symposium on Low Electronics and Design (ISLPED'01), Huntington Beach, CA, United States.
Chung, E. Y. ; Benini, L. ; De Micheli, G. / Automatic source code specialization for energy reduction. Paper presented at International Symposium on Low Electronics and Design (ISLPED'01), Huntington Beach, CA, United States.4 p.
@conference{516ce20682d842ff8eeccd45f9e271df,
title = "Automatic source code specialization for energy reduction",
abstract = "This paper presents a framework to reduce the computational effort of software programs, using value profiling and partial evaluation. Our tool reduces computational effort by specializing a program for highly expected situations and such a reduction translates into both energy and performance improvement. Procedure calls executed frequently with same parameter values are defined as highly expected situations (common cases). The choice of the best transformation of common cases is achieved by solving three search problems. The first identifies effective common cases to be specialized, the second searches for an optimal solution for effective common case, and the third examines the interplay among the specialized cases. Our technique improves both energy consumption and performance of the source code up to more than twice and in average about 25{\%} over the original program. Also, our pruning techniques reduce the searching time by 80{\%} compared to exhaustive approach.",
author = "Chung, {E. Y.} and L. Benini and {De Micheli}, G.",
year = "2001",
month = "1",
day = "1",
language = "English",
pages = "80--83",
note = "International Symposium on Low Electronics and Design (ISLPED'01) ; Conference date: 06-08-2001 Through 07-08-2001",

}

Chung, EY, Benini, L & De Micheli, G 2001, 'Automatic source code specialization for energy reduction' Paper presented at International Symposium on Low Electronics and Design (ISLPED'01), Huntington Beach, CA, United States, 01/8/6 - 01/8/7, pp. 80-83.

Automatic source code specialization for energy reduction. / Chung, E. Y.; Benini, L.; De Micheli, G.

2001. 80-83 Paper presented at International Symposium on Low Electronics and Design (ISLPED'01), Huntington Beach, CA, United States.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Automatic source code specialization for energy reduction

AU - Chung, E. Y.

AU - Benini, L.

AU - De Micheli, G.

PY - 2001/1/1

Y1 - 2001/1/1

N2 - This paper presents a framework to reduce the computational effort of software programs, using value profiling and partial evaluation. Our tool reduces computational effort by specializing a program for highly expected situations and such a reduction translates into both energy and performance improvement. Procedure calls executed frequently with same parameter values are defined as highly expected situations (common cases). The choice of the best transformation of common cases is achieved by solving three search problems. The first identifies effective common cases to be specialized, the second searches for an optimal solution for effective common case, and the third examines the interplay among the specialized cases. Our technique improves both energy consumption and performance of the source code up to more than twice and in average about 25% over the original program. Also, our pruning techniques reduce the searching time by 80% compared to exhaustive approach.

AB - This paper presents a framework to reduce the computational effort of software programs, using value profiling and partial evaluation. Our tool reduces computational effort by specializing a program for highly expected situations and such a reduction translates into both energy and performance improvement. Procedure calls executed frequently with same parameter values are defined as highly expected situations (common cases). The choice of the best transformation of common cases is achieved by solving three search problems. The first identifies effective common cases to be specialized, the second searches for an optimal solution for effective common case, and the third examines the interplay among the specialized cases. Our technique improves both energy consumption and performance of the source code up to more than twice and in average about 25% over the original program. Also, our pruning techniques reduce the searching time by 80% compared to exhaustive approach.

UR - http://www.scopus.com/inward/record.url?scp=0034876764&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034876764&partnerID=8YFLogxK

M3 - Paper

SP - 80

EP - 83

ER -

Chung EY, Benini L, De Micheli G. Automatic source code specialization for energy reduction. 2001. Paper presented at International Symposium on Low Electronics and Design (ISLPED'01), Huntington Beach, CA, United States.