Compile-time based performance prediction

Calin Cascaval; Luiz De Rose; David A. Padua; Daniel A. Reed

doi:10.1007/3-540-44905-1_23

Compile-time based performance prediction

Calin Cascaval, Luiz De Rose, David A. Padua, Daniel A. Reed

Siebel School of Computing and Data Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper we present results we obtained using a compiler to predict performance of scientific codes. The compiler, Polaris [3], is both the primary tool for estimating the performance of a range of codes, and the beneficiary of the results obtained from predicting the program behavior at compile time. We show that a simple compile-time model, augmented with profiling data obtained using very light instrumentation, can be accurate within 20% (on average) of the measured performance for codes using both dense and sparse computational methods.

Original language	English (US)
Title of host publication	Languages and Compilers for Parallel Computing - 12th International Workshop, LCPC 1999, Proceedings
Editors	Larry Carter, Jeanne Ferrante
Publisher	Springer
Pages	365-379
Number of pages	15
ISBN (Print)	9783540678588
DOIs	https://doi.org/10.1007/3-540-44905-1_23
State	Published - 2000
Event	12th International Workshop on Languages and Compilers for Parallel Computing, LCPC 1999 - La Jolla, United States Duration: Aug 4 1999 → Aug 6 1999

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	1863
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	12th International Workshop on Languages and Compilers for Parallel Computing, LCPC 1999
Country/Territory	United States
City	La Jolla
Period	8/4/99 → 8/6/99

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Online availability

10.1007/3-540-44905-1_23

Library availability

Discover UIUC Full Text

Cite this

Cascaval, C., De Rose, L., Padua, D. A., & Reed, D. A. (2000). Compile-time based performance prediction. In L. Carter, & J. Ferrante (Eds.), Languages and Compilers for Parallel Computing - 12th International Workshop, LCPC 1999, Proceedings (pp. 365-379). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1863). Springer. https://doi.org/10.1007/3-540-44905-1_23

Compile-time based performance prediction. / Cascaval, Calin; De Rose, Luiz; Padua, David A. et al.
Languages and Compilers for Parallel Computing - 12th International Workshop, LCPC 1999, Proceedings. ed. / Larry Carter; Jeanne Ferrante. Springer, 2000. p. 365-379 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 1863).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Cascaval, C, De Rose, L, Padua, DA & Reed, DA 2000, Compile-time based performance prediction. in L Carter & J Ferrante (eds), Languages and Compilers for Parallel Computing - 12th International Workshop, LCPC 1999, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 1863, Springer, pp. 365-379, 12th International Workshop on Languages and Compilers for Parallel Computing, LCPC 1999, La Jolla, United States, 8/4/99. https://doi.org/10.1007/3-540-44905-1_23

Cascaval C, De Rose L, Padua DA, Reed DA. Compile-time based performance prediction. In Carter L, Ferrante J, editors, Languages and Compilers for Parallel Computing - 12th International Workshop, LCPC 1999, Proceedings. Springer. 2000. p. 365-379. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/3-540-44905-1_23

Cascaval, Calin ; De Rose, Luiz ; Padua, David A. et al. / Compile-time based performance prediction. Languages and Compilers for Parallel Computing - 12th International Workshop, LCPC 1999, Proceedings. editor / Larry Carter ; Jeanne Ferrante. Springer, 2000. pp. 365-379 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{cbfa441828284e109006ca0666632e8c,

title = "Compile-time based performance prediction",

abstract = "In this paper we present results we obtained using a compiler to predict performance of scientific codes. The compiler, Polaris [3], is both the primary tool for estimating the performance of a range of codes, and the beneficiary of the results obtained from predicting the program behavior at compile time. We show that a simple compile-time model, augmented with profiling data obtained using very light instrumentation, can be accurate within 20% (on average) of the measured performance for codes using both dense and sparse computational methods.",

author = "Calin Cascaval and {De Rose}, Luiz and Padua, {David A.} and Reed, {Daniel A.}",

note = "Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 2000.; 12th International Workshop on Languages and Compilers for Parallel Computing, LCPC 1999 ; Conference date: 04-08-1999 Through 06-08-1999",

year = "2000",

doi = "10.1007/3-540-44905-1_23",

language = "English (US)",

isbn = "9783540678588",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "365--379",

editor = "Larry Carter and Jeanne Ferrante",

booktitle = "Languages and Compilers for Parallel Computing - 12th International Workshop, LCPC 1999, Proceedings",

address = "Germany",

}

TY - GEN

T1 - Compile-time based performance prediction

AU - Cascaval, Calin

AU - De Rose, Luiz

AU - Padua, David A.

AU - Reed, Daniel A.

PY - 2000

Y1 - 2000

N2 - In this paper we present results we obtained using a compiler to predict performance of scientific codes. The compiler, Polaris [3], is both the primary tool for estimating the performance of a range of codes, and the beneficiary of the results obtained from predicting the program behavior at compile time. We show that a simple compile-time model, augmented with profiling data obtained using very light instrumentation, can be accurate within 20% (on average) of the measured performance for codes using both dense and sparse computational methods.

AB - In this paper we present results we obtained using a compiler to predict performance of scientific codes. The compiler, Polaris [3], is both the primary tool for estimating the performance of a range of codes, and the beneficiary of the results obtained from predicting the program behavior at compile time. We show that a simple compile-time model, augmented with profiling data obtained using very light instrumentation, can be accurate within 20% (on average) of the measured performance for codes using both dense and sparse computational methods.

UR - http://www.scopus.com/inward/record.url?scp=84948974925&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948974925&partnerID=8YFLogxK

U2 - 10.1007/3-540-44905-1_23

DO - 10.1007/3-540-44905-1_23

M3 - Conference contribution

AN - SCOPUS:84948974925

SN - 9783540678588

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 365

EP - 379

BT - Languages and Compilers for Parallel Computing - 12th International Workshop, LCPC 1999, Proceedings

A2 - Carter, Larry

A2 - Ferrante, Jeanne

PB - Springer

T2 - 12th International Workshop on Languages and Compilers for Parallel Computing, LCPC 1999

Y2 - 4 August 1999 through 6 August 1999

ER -

Compile-time based performance prediction

Abstract

Publication series

Other

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this