Using unfolding to minimize inter-iteration dependencies

Timothy W. O'Neil; Edwin H.M. Sha

Using unfolding to minimize inter-iteration dependencies

Research output: Contribution to journal › Conference article › peer-review

Abstract

Since data dependencies greatly decrease instruction level parallelism, minimizing dependencies becomes a crucial part of the process of parallelizing sequential code. Eliminating all unnecessary hazards leads to the more efficient use of resources, fewer processor stalls and easily maintainable code. Previously we proposed a novel approach for eliminating redundant data dependencies from code. In this paper, we review this method and show how this elimination technique may be combined with unfolding so as to parallelize code even further.

Original language	English
Article number	439-186
Pages (from-to)	342-347
Number of pages	6
Journal	Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems
Volume	16
State	Published - 2004
Externally published	Yes
Event	Proceedings of the 16th IASTED International Conference on Parallel and Distributed Computing and Systems - Cambridge, MA, United States Duration: 9 Nov 2004 → 11 Nov 2004

Keywords

Compiler optimization
Data dependence analysis
Redundant dependence analysis
Unfolding

Cite this

@article{55f0947d3fdc4af7a2d21b108ce9ed58,

title = "Using unfolding to minimize inter-iteration dependencies",

abstract = "Since data dependencies greatly decrease instruction level parallelism, minimizing dependencies becomes a crucial part of the process of parallelizing sequential code. Eliminating all unnecessary hazards leads to the more efficient use of resources, fewer processor stalls and easily maintainable code. Previously we proposed a novel approach for eliminating redundant data dependencies from code. In this paper, we review this method and show how this elimination technique may be combined with unfolding so as to parallelize code even further.",

keywords = "Compiler optimization, Data dependence analysis, Redundant dependence analysis, Unfolding",

author = "O'Neil, \{Timothy W.\} and Sha, \{Edwin H.M.\}",

year = "2004",

language = "英语",

volume = "16",

pages = "342--347",

journal = "Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems",

issn = "1027-2658",

publisher = "Acta Press",

note = "Proceedings of the 16th IASTED International Conference on Parallel and Distributed Computing and Systems ; Conference date: 09-11-2004 Through 11-11-2004",

}

TY - JOUR

T1 - Using unfolding to minimize inter-iteration dependencies

AU - O'Neil, Timothy W.

AU - Sha, Edwin H.M.

PY - 2004

Y1 - 2004

N2 - Since data dependencies greatly decrease instruction level parallelism, minimizing dependencies becomes a crucial part of the process of parallelizing sequential code. Eliminating all unnecessary hazards leads to the more efficient use of resources, fewer processor stalls and easily maintainable code. Previously we proposed a novel approach for eliminating redundant data dependencies from code. In this paper, we review this method and show how this elimination technique may be combined with unfolding so as to parallelize code even further.

AB - Since data dependencies greatly decrease instruction level parallelism, minimizing dependencies becomes a crucial part of the process of parallelizing sequential code. Eliminating all unnecessary hazards leads to the more efficient use of resources, fewer processor stalls and easily maintainable code. Previously we proposed a novel approach for eliminating redundant data dependencies from code. In this paper, we review this method and show how this elimination technique may be combined with unfolding so as to parallelize code even further.

KW - Compiler optimization

KW - Data dependence analysis

KW - Redundant dependence analysis

KW - Unfolding

UR - https://www.scopus.com/pages/publications/11844294893

M3 - 会议文章

AN - SCOPUS:11844294893

SN - 1027-2658

VL - 16

SP - 342

EP - 347

JO - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems

JF - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems

M1 - 439-186

T2 - Proceedings of the 16th IASTED International Conference on Parallel and Distributed Computing and Systems

Y2 - 9 November 2004 through 11 November 2004

ER -

Using unfolding to minimize inter-iteration dependencies

Abstract

Keywords

Other files and links

Fingerprint

Cite this