HELIX-RC: An architecture-compiler co-design for automatic parallelization of irregular programs

Simone Campanoni; Kevin Brownell; Svilen Kanev; Timothy M. Jones; Gu Yeon Wei; David Brooks

doi:10.1109/ISCA.2014.6853215

HELIX-RC: An architecture-compiler co-design for automatic parallelization of irregular programs

Simone Campanoni, Kevin Brownell, Svilen Kanev, Timothy M. Jones, Gu Yeon Wei, David Brooks

Computer Science

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

41 Scopus citations

Abstract

Data dependences in sequential programs limit parallelization because extracted threads cannot run independently. Although thread-level speculation can avoid the need for precise dependence analysis, communication overheads required to synchronize actual dependences counteract the benefits of parallelization. To address these challenges, we propose a lightweight architectural enhancement co-designed with a parallelizing compiler, which together can decouple communication from thread execution. Simulations of these approaches, applied to a processor with 16 Intel Atom-like cores, show an average of 6.85× performance speedup for six SPEC CINT2000 benchmarks.

Original language	English (US)
Title of host publication	41st Annual International Symposium on Computer Architecture, ISCA 2014 - Conference Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	217-228
Number of pages	12
ISBN (Print)	9781479943968
DOIs	https://doi.org/10.1109/ISCA.2014.6853215
State	Published - 2014
Event	2014 ACM/IEEE 41st International Symposium on Computer Architecture, ISCA 2014 - Minneapolis, MN, United States Duration: Jun 14 2014 → Jun 18 2014

Publication series

Name	Proceedings - International Symposium on Computer Architecture
ISSN (Print)	1063-6897

Other

Other	2014 ACM/IEEE 41st International Symposium on Computer Architecture, ISCA 2014
Country/Territory	United States
City	Minneapolis, MN
Period	6/14/14 → 6/18/14

ASJC Scopus subject areas

Hardware and Architecture

Access to Document

10.1109/ISCA.2014.6853215

Cite this

Campanoni, S., Brownell, K., Kanev, S., Jones, T. M., Wei, G. Y., & Brooks, D. (2014). HELIX-RC: An architecture-compiler co-design for automatic parallelization of irregular programs. In 41st Annual International Symposium on Computer Architecture, ISCA 2014 - Conference Proceedings (pp. 217-228). Article 6853215 (Proceedings - International Symposium on Computer Architecture). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISCA.2014.6853215

Campanoni, Simone ; Brownell, Kevin ; Kanev, Svilen et al. / HELIX-RC : An architecture-compiler co-design for automatic parallelization of irregular programs. 41st Annual International Symposium on Computer Architecture, ISCA 2014 - Conference Proceedings. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 217-228 (Proceedings - International Symposium on Computer Architecture).

@inproceedings{9411de3e5d8e41e0bb0f5ac78a076f18,

title = "HELIX-RC: An architecture-compiler co-design for automatic parallelization of irregular programs",

abstract = "Data dependences in sequential programs limit parallelization because extracted threads cannot run independently. Although thread-level speculation can avoid the need for precise dependence analysis, communication overheads required to synchronize actual dependences counteract the benefits of parallelization. To address these challenges, we propose a lightweight architectural enhancement co-designed with a parallelizing compiler, which together can decouple communication from thread execution. Simulations of these approaches, applied to a processor with 16 Intel Atom-like cores, show an average of 6.85× performance speedup for six SPEC CINT2000 benchmarks.",

author = "Simone Campanoni and Kevin Brownell and Svilen Kanev and Jones, {Timothy M.} and Wei, {Gu Yeon} and David Brooks",

year = "2014",

doi = "10.1109/ISCA.2014.6853215",

language = "English (US)",

isbn = "9781479943968",

series = "Proceedings - International Symposium on Computer Architecture",

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

pages = "217--228",

booktitle = "41st Annual International Symposium on Computer Architecture, ISCA 2014 - Conference Proceedings",

address = "United States",

note = "2014 ACM/IEEE 41st International Symposium on Computer Architecture, ISCA 2014 ; Conference date: 14-06-2014 Through 18-06-2014",

}

Campanoni, S, Brownell, K, Kanev, S, Jones, TM, Wei, GY & Brooks, D 2014, HELIX-RC: An architecture-compiler co-design for automatic parallelization of irregular programs. in 41st Annual International Symposium on Computer Architecture, ISCA 2014 - Conference Proceedings., 6853215, Proceedings - International Symposium on Computer Architecture, Institute of Electrical and Electronics Engineers Inc., pp. 217-228, 2014 ACM/IEEE 41st International Symposium on Computer Architecture, ISCA 2014, Minneapolis, MN, United States, 6/14/14. https://doi.org/10.1109/ISCA.2014.6853215

HELIX-RC: An architecture-compiler co-design for automatic parallelization of irregular programs. / Campanoni, Simone; Brownell, Kevin; Kanev, Svilen et al.
41st Annual International Symposium on Computer Architecture, ISCA 2014 - Conference Proceedings. Institute of Electrical and Electronics Engineers Inc., 2014. p. 217-228 6853215 (Proceedings - International Symposium on Computer Architecture).

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

TY - GEN

T1 - HELIX-RC

T2 - 2014 ACM/IEEE 41st International Symposium on Computer Architecture, ISCA 2014

AU - Campanoni, Simone

AU - Brownell, Kevin

AU - Kanev, Svilen

AU - Jones, Timothy M.

AU - Wei, Gu Yeon

AU - Brooks, David

PY - 2014

Y1 - 2014

N2 - Data dependences in sequential programs limit parallelization because extracted threads cannot run independently. Although thread-level speculation can avoid the need for precise dependence analysis, communication overheads required to synchronize actual dependences counteract the benefits of parallelization. To address these challenges, we propose a lightweight architectural enhancement co-designed with a parallelizing compiler, which together can decouple communication from thread execution. Simulations of these approaches, applied to a processor with 16 Intel Atom-like cores, show an average of 6.85× performance speedup for six SPEC CINT2000 benchmarks.

AB - Data dependences in sequential programs limit parallelization because extracted threads cannot run independently. Although thread-level speculation can avoid the need for precise dependence analysis, communication overheads required to synchronize actual dependences counteract the benefits of parallelization. To address these challenges, we propose a lightweight architectural enhancement co-designed with a parallelizing compiler, which together can decouple communication from thread execution. Simulations of these approaches, applied to a processor with 16 Intel Atom-like cores, show an average of 6.85× performance speedup for six SPEC CINT2000 benchmarks.

UR - http://www.scopus.com/inward/record.url?scp=84905455779&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84905455779&partnerID=8YFLogxK

U2 - 10.1109/ISCA.2014.6853215

DO - 10.1109/ISCA.2014.6853215

M3 - Conference contribution

AN - SCOPUS:84905455779

SN - 9781479943968

T3 - Proceedings - International Symposium on Computer Architecture

SP - 217

EP - 228

BT - 41st Annual International Symposium on Computer Architecture, ISCA 2014 - Conference Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 14 June 2014 through 18 June 2014

ER -

Campanoni S, Brownell K, Kanev S, Jones TM, Wei GY, Brooks D. HELIX-RC: An architecture-compiler co-design for automatic parallelization of irregular programs. In 41st Annual International Symposium on Computer Architecture, ISCA 2014 - Conference Proceedings. Institute of Electrical and Electronics Engineers Inc. 2014. p. 217-228. 6853215. (Proceedings - International Symposium on Computer Architecture). doi: 10.1109/ISCA.2014.6853215

HELIX-RC: An architecture-compiler co-design for automatic parallelization of irregular programs

Abstract

Publication series

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this