Restructuring programs for high-speed computers with Polaris

Bill Blume; Rudolf Eigenmann; Keith Faigin; John Grout; Jaejin Lee; Tom Lawrence; Jay Hoeflinger; David Padua; Yunheung Paek; Paul Petersen; Bill Pottenger; Lawrence Rauchwerger; Peng Tu; Stephen Weatherford

doi:10.1109/ICPPW.1996.538601

Restructuring programs for high-speed computers with Polaris

Bill Blume, Rudolf Eigenmann, Keith Faigin, John Grout, Jaejin Lee, Tom Lawrence, Jay Hoeflinger, David Padua, Yunheung Paek, Paul Petersen, Bill Pottenger, Lawrence Rauchwerger, Peng Tu, Stephen Weatherford

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

Abstract

The ability to automatically parallelize standard programming languages results in program portability across a wide range of machine architectures. It is the goal of the Polaris project to develop a new parallelizing compiler that overcomes limitations of current compilers. While current parallelizing compilers may succeed on small kernels, they often fail to extract any meaningful parallelism from whole applications. After a study of application codes, it was concluded that by adding a few new techniques to current compilers, automatic parallelization becomes feasible for a range of whole applications. The techniques needed are interprocedural analysis, scalar and array privatization, symbolic dependence analysis, and advanced induction and reduction recognition and elimination, along with run-time techniques to permit the parallelization of loops with unknown dependence relations.

Original language	English (US)
Title of host publication	Proceedings of the 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996
Editors	H.J. Siegel
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	149-161
Number of pages	13
ISBN (Electronic)	081867623X
DOIs	https://doi.org/10.1109/ICPPW.1996.538601
State	Published - 1996
Event	1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996 - Ithaca, United States Duration: Aug 12 1996 → Aug 12 1996

Publication series

Name	Proceedings of the International Conference on Parallel Processing Workshops
ISSN (Print)	1530-2016

Other

Other	1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996
Country/Territory	United States
City	Ithaca
Period	8/12/96 → 8/12/96

ASJC Scopus subject areas

Software
General Mathematics
Hardware and Architecture

Online availability

10.1109/ICPPW.1996.538601

Library availability

Discover UIUC Full Text

Cite this

Blume, B., Eigenmann, R., Faigin, K., Grout, J., Lee, J., Lawrence, T., Hoeflinger, J., Padua, D., Paek, Y., Petersen, P., Pottenger, B., Rauchwerger, L., Tu, P., & Weatherford, S. (1996). Restructuring programs for high-speed computers with Polaris. In H. J. Siegel (Ed.), Proceedings of the 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996 (pp. 149-161). Article 538601 (Proceedings of the International Conference on Parallel Processing Workshops). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICPPW.1996.538601

Restructuring programs for high-speed computers with Polaris. / Blume, Bill; Eigenmann, Rudolf; Faigin, Keith et al.
Proceedings of the 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996. ed. / H.J. Siegel. Institute of Electrical and Electronics Engineers Inc., 1996. p. 149-161 538601 (Proceedings of the International Conference on Parallel Processing Workshops).

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

Blume, B, Eigenmann, R, Faigin, K, Grout, J, Lee, J, Lawrence, T, Hoeflinger, J, Padua, D, Paek, Y, Petersen, P, Pottenger, B, Rauchwerger, L, Tu, P & Weatherford, S 1996, Restructuring programs for high-speed computers with Polaris. in HJ Siegel (ed.), Proceedings of the 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996., 538601, Proceedings of the International Conference on Parallel Processing Workshops, Institute of Electrical and Electronics Engineers Inc., pp. 149-161, 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996, Ithaca, United States, 8/12/96. https://doi.org/10.1109/ICPPW.1996.538601

Blume B, Eigenmann R, Faigin K, Grout J, Lee J, Lawrence T et al. Restructuring programs for high-speed computers with Polaris. In Siegel HJ, editor, Proceedings of the 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996. Institute of Electrical and Electronics Engineers Inc. 1996. p. 149-161. 538601. (Proceedings of the International Conference on Parallel Processing Workshops). doi: 10.1109/ICPPW.1996.538601

Blume, Bill ; Eigenmann, Rudolf ; Faigin, Keith et al. / Restructuring programs for high-speed computers with Polaris. Proceedings of the 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996. editor / H.J. Siegel. Institute of Electrical and Electronics Engineers Inc., 1996. pp. 149-161 (Proceedings of the International Conference on Parallel Processing Workshops).

@inproceedings{c6251ecb318b40fd94fa763b56b6c4a2,

title = "Restructuring programs for high-speed computers with Polaris",

abstract = "The ability to automatically parallelize standard programming languages results in program portability across a wide range of machine architectures. It is the goal of the Polaris project to develop a new parallelizing compiler that overcomes limitations of current compilers. While current parallelizing compilers may succeed on small kernels, they often fail to extract any meaningful parallelism from whole applications. After a study of application codes, it was concluded that by adding a few new techniques to current compilers, automatic parallelization becomes feasible for a range of whole applications. The techniques needed are interprocedural analysis, scalar and array privatization, symbolic dependence analysis, and advanced induction and reduction recognition and elimination, along with run-time techniques to permit the parallelization of loops with unknown dependence relations.",

author = "Bill Blume and Rudolf Eigenmann and Keith Faigin and John Grout and Jaejin Lee and Tom Lawrence and Jay Hoeflinger and David Padua and Yunheung Paek and Paul Petersen and Bill Pottenger and Lawrence Rauchwerger and Peng Tu and Stephen Weatherford",

note = "Publisher Copyright: {\textcopyright} 1996 IEEE.; 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996 ; Conference date: 12-08-1996 Through 12-08-1996",

year = "1996",

doi = "10.1109/ICPPW.1996.538601",

language = "English (US)",

series = "Proceedings of the International Conference on Parallel Processing Workshops",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "149--161",

editor = "H.J. Siegel",

booktitle = "Proceedings of the 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996",

address = "United States",

}

TY - GEN

T1 - Restructuring programs for high-speed computers with Polaris

AU - Blume, Bill

AU - Eigenmann, Rudolf

AU - Faigin, Keith

AU - Grout, John

AU - Lee, Jaejin

AU - Lawrence, Tom

AU - Hoeflinger, Jay

AU - Padua, David

AU - Paek, Yunheung

AU - Petersen, Paul

AU - Pottenger, Bill

AU - Rauchwerger, Lawrence

AU - Tu, Peng

AU - Weatherford, Stephen

PY - 1996

Y1 - 1996

N2 - The ability to automatically parallelize standard programming languages results in program portability across a wide range of machine architectures. It is the goal of the Polaris project to develop a new parallelizing compiler that overcomes limitations of current compilers. While current parallelizing compilers may succeed on small kernels, they often fail to extract any meaningful parallelism from whole applications. After a study of application codes, it was concluded that by adding a few new techniques to current compilers, automatic parallelization becomes feasible for a range of whole applications. The techniques needed are interprocedural analysis, scalar and array privatization, symbolic dependence analysis, and advanced induction and reduction recognition and elimination, along with run-time techniques to permit the parallelization of loops with unknown dependence relations.

AB - The ability to automatically parallelize standard programming languages results in program portability across a wide range of machine architectures. It is the goal of the Polaris project to develop a new parallelizing compiler that overcomes limitations of current compilers. While current parallelizing compilers may succeed on small kernels, they often fail to extract any meaningful parallelism from whole applications. After a study of application codes, it was concluded that by adding a few new techniques to current compilers, automatic parallelization becomes feasible for a range of whole applications. The techniques needed are interprocedural analysis, scalar and array privatization, symbolic dependence analysis, and advanced induction and reduction recognition and elimination, along with run-time techniques to permit the parallelization of loops with unknown dependence relations.

UR - http://www.scopus.com/inward/record.url?scp=0000615819&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0000615819&partnerID=8YFLogxK

U2 - 10.1109/ICPPW.1996.538601

DO - 10.1109/ICPPW.1996.538601

M3 - Conference contribution

AN - SCOPUS:0000615819

T3 - Proceedings of the International Conference on Parallel Processing Workshops

SP - 149

EP - 161

BT - Proceedings of the 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996

A2 - Siegel, H.J.

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1996 ICPP Workshop on Challenges for Parallel Processing, ICPPW 1996

Y2 - 12 August 1996 through 12 August 1996

ER -

Restructuring programs for high-speed computers with Polaris

Abstract

Publication series

Other

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this