Improving spatial locality of programs via data mining

Karlton Sequeira, Mohammed Zaki, Boleslaw Szymanski, Christopher Carothers

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In most computer systems, page fault rate is currently minimized by generic page replacement algorithms which try to model the temporal locality inherent in programs. In this paper, we propose two algorithms, one greedy and the other stochastic, designed for program specific code restructuring as a means of increasing spatial locality within a program. Both algorithms effectively decrease average working set size and hence the page fault rate. Our methods are more effective than traditional approaches due to use of domain information. We illustrate the efficacy of our algorithms on actual data mining algorithms.

Original languageEnglish
Title of host publicationProceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
Pages649-654
Number of pages6
DOIs
Publication statusPublished - 1 Dec 2003
Externally publishedYes
Event9th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '03 - Washington, DC, United States
Duration: 24 Aug 200327 Aug 2003

Other

Other9th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '03
CountryUnited States
CityWashington, DC
Period24/8/0327/8/03

Fingerprint

Data mining
Computer systems

Keywords

  • Code restructuring
  • Page clustering
  • Program locality

ASJC Scopus subject areas

  • Software
  • Information Systems

Cite this

Sequeira, K., Zaki, M., Szymanski, B., & Carothers, C. (2003). Improving spatial locality of programs via data mining. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 649-654) https://doi.org/10.1145/956750.956834

Improving spatial locality of programs via data mining. / Sequeira, Karlton; Zaki, Mohammed; Szymanski, Boleslaw; Carothers, Christopher.

Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2003. p. 649-654.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Sequeira, K, Zaki, M, Szymanski, B & Carothers, C 2003, Improving spatial locality of programs via data mining. in Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. pp. 649-654, 9th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '03, Washington, DC, United States, 24/8/03. https://doi.org/10.1145/956750.956834
Sequeira K, Zaki M, Szymanski B, Carothers C. Improving spatial locality of programs via data mining. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2003. p. 649-654 https://doi.org/10.1145/956750.956834
Sequeira, Karlton ; Zaki, Mohammed ; Szymanski, Boleslaw ; Carothers, Christopher. / Improving spatial locality of programs via data mining. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2003. pp. 649-654
@inproceedings{eb32cfa0f546412997019f90055b8e61,
title = "Improving spatial locality of programs via data mining",
abstract = "In most computer systems, page fault rate is currently minimized by generic page replacement algorithms which try to model the temporal locality inherent in programs. In this paper, we propose two algorithms, one greedy and the other stochastic, designed for program specific code restructuring as a means of increasing spatial locality within a program. Both algorithms effectively decrease average working set size and hence the page fault rate. Our methods are more effective than traditional approaches due to use of domain information. We illustrate the efficacy of our algorithms on actual data mining algorithms.",
keywords = "Code restructuring, Page clustering, Program locality",
author = "Karlton Sequeira and Mohammed Zaki and Boleslaw Szymanski and Christopher Carothers",
year = "2003",
month = "12",
day = "1",
doi = "10.1145/956750.956834",
language = "English",
pages = "649--654",
booktitle = "Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining",

}

TY - GEN

T1 - Improving spatial locality of programs via data mining

AU - Sequeira, Karlton

AU - Zaki, Mohammed

AU - Szymanski, Boleslaw

AU - Carothers, Christopher

PY - 2003/12/1

Y1 - 2003/12/1

N2 - In most computer systems, page fault rate is currently minimized by generic page replacement algorithms which try to model the temporal locality inherent in programs. In this paper, we propose two algorithms, one greedy and the other stochastic, designed for program specific code restructuring as a means of increasing spatial locality within a program. Both algorithms effectively decrease average working set size and hence the page fault rate. Our methods are more effective than traditional approaches due to use of domain information. We illustrate the efficacy of our algorithms on actual data mining algorithms.

AB - In most computer systems, page fault rate is currently minimized by generic page replacement algorithms which try to model the temporal locality inherent in programs. In this paper, we propose two algorithms, one greedy and the other stochastic, designed for program specific code restructuring as a means of increasing spatial locality within a program. Both algorithms effectively decrease average working set size and hence the page fault rate. Our methods are more effective than traditional approaches due to use of domain information. We illustrate the efficacy of our algorithms on actual data mining algorithms.

KW - Code restructuring

KW - Page clustering

KW - Program locality

UR - http://www.scopus.com/inward/record.url?scp=77952335355&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77952335355&partnerID=8YFLogxK

U2 - 10.1145/956750.956834

DO - 10.1145/956750.956834

M3 - Conference contribution

AN - SCOPUS:77952335355

SP - 649

EP - 654

BT - Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

ER -