Reference distance as a metric for data locality

Changwoo Pyo, Kyung Ho Lee, Kyung Woo Lee, Hye Kyung Han

Research output: Chapter in Book/Report/Conference proceedingChapter

11 Citations (Scopus)

Abstract

We propose reference distance as a metric for data locality. Reference distance is the number of referenced memory blocks between two successive references to the same memory block. Effectiveness of program transformations for data locality can be evaluated by using reference distance. Our claim is demonstrated by showing the change in data locality and speedup of matrix multiplication due to loop interchange, loop tiling, and loop unrolling. We also present a result of an experiment using loop distribution and unrolling with respect to two subroutines of Perfect benchmark programs. Reference distance would serve as a useful tool for developing program transformation scheme for data locality optimization.

Original languageEnglish
Title of host publicationProceedings of the Conference on High Performance Computing on the Information Superhighway, HPC Asia'97
Editors Anon
Place of PublicationLos Alamitos, CA, United States
PublisherIEEE
Pages151-156
Number of pages6
Publication statusPublished - 1997 Jan 1
Externally publishedYes
EventProceedings of the 1997 2nd High Performance Computing on the Information Superhighway, HPC Asia'97 - Seoul, South Korea
Duration: 1997 Apr 281997 May 2

Other

OtherProceedings of the 1997 2nd High Performance Computing on the Information Superhighway, HPC Asia'97
CitySeoul, South Korea
Period97/4/2897/5/2

Fingerprint

Data storage equipment
Subroutines
Interchanges
Experiments

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Pyo, C., Lee, K. H., Lee, K. W., & Han, H. K. (1997). Reference distance as a metric for data locality. In Anon (Ed.), Proceedings of the Conference on High Performance Computing on the Information Superhighway, HPC Asia'97 (pp. 151-156). Los Alamitos, CA, United States: IEEE.

Reference distance as a metric for data locality. / Pyo, Changwoo; Lee, Kyung Ho; Lee, Kyung Woo; Han, Hye Kyung.

Proceedings of the Conference on High Performance Computing on the Information Superhighway, HPC Asia'97. ed. / Anon. Los Alamitos, CA, United States : IEEE, 1997. p. 151-156.

Research output: Chapter in Book/Report/Conference proceedingChapter

Pyo, C, Lee, KH, Lee, KW & Han, HK 1997, Reference distance as a metric for data locality. in Anon (ed.), Proceedings of the Conference on High Performance Computing on the Information Superhighway, HPC Asia'97. IEEE, Los Alamitos, CA, United States, pp. 151-156, Proceedings of the 1997 2nd High Performance Computing on the Information Superhighway, HPC Asia'97, Seoul, South Korea, 97/4/28.
Pyo C, Lee KH, Lee KW, Han HK. Reference distance as a metric for data locality. In Anon, editor, Proceedings of the Conference on High Performance Computing on the Information Superhighway, HPC Asia'97. Los Alamitos, CA, United States: IEEE. 1997. p. 151-156
Pyo, Changwoo ; Lee, Kyung Ho ; Lee, Kyung Woo ; Han, Hye Kyung. / Reference distance as a metric for data locality. Proceedings of the Conference on High Performance Computing on the Information Superhighway, HPC Asia'97. editor / Anon. Los Alamitos, CA, United States : IEEE, 1997. pp. 151-156
@inbook{d0918a85eeb74f9eaf199143e0f4a253,
title = "Reference distance as a metric for data locality",
abstract = "We propose reference distance as a metric for data locality. Reference distance is the number of referenced memory blocks between two successive references to the same memory block. Effectiveness of program transformations for data locality can be evaluated by using reference distance. Our claim is demonstrated by showing the change in data locality and speedup of matrix multiplication due to loop interchange, loop tiling, and loop unrolling. We also present a result of an experiment using loop distribution and unrolling with respect to two subroutines of Perfect benchmark programs. Reference distance would serve as a useful tool for developing program transformation scheme for data locality optimization.",
author = "Changwoo Pyo and Lee, {Kyung Ho} and Lee, {Kyung Woo} and Han, {Hye Kyung}",
year = "1997",
month = "1",
day = "1",
language = "English",
pages = "151--156",
editor = "Anon",
booktitle = "Proceedings of the Conference on High Performance Computing on the Information Superhighway, HPC Asia'97",
publisher = "IEEE",

}

TY - CHAP

T1 - Reference distance as a metric for data locality

AU - Pyo, Changwoo

AU - Lee, Kyung Ho

AU - Lee, Kyung Woo

AU - Han, Hye Kyung

PY - 1997/1/1

Y1 - 1997/1/1

N2 - We propose reference distance as a metric for data locality. Reference distance is the number of referenced memory blocks between two successive references to the same memory block. Effectiveness of program transformations for data locality can be evaluated by using reference distance. Our claim is demonstrated by showing the change in data locality and speedup of matrix multiplication due to loop interchange, loop tiling, and loop unrolling. We also present a result of an experiment using loop distribution and unrolling with respect to two subroutines of Perfect benchmark programs. Reference distance would serve as a useful tool for developing program transformation scheme for data locality optimization.

AB - We propose reference distance as a metric for data locality. Reference distance is the number of referenced memory blocks between two successive references to the same memory block. Effectiveness of program transformations for data locality can be evaluated by using reference distance. Our claim is demonstrated by showing the change in data locality and speedup of matrix multiplication due to loop interchange, loop tiling, and loop unrolling. We also present a result of an experiment using loop distribution and unrolling with respect to two subroutines of Perfect benchmark programs. Reference distance would serve as a useful tool for developing program transformation scheme for data locality optimization.

UR - http://www.scopus.com/inward/record.url?scp=0030649712&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030649712&partnerID=8YFLogxK

M3 - Chapter

SP - 151

EP - 156

BT - Proceedings of the Conference on High Performance Computing on the Information Superhighway, HPC Asia'97

A2 - Anon, null

PB - IEEE

CY - Los Alamitos, CA, United States

ER -