Two-dimensional data locality: Definition, abstraction, and application

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

2 Citations (Scopus)

Abstract

Data locality has been a central theme in the compiler optimization world for a long time. Most of the prior compiler techniques try to optimize data locality in a one-dimensional linear address space. However, there are many problems out there where the domain for data locality can be two or higher dimensional. For example, in a two-dimensional mesh network environment, each node has connections with its four neighbors, and therefore, the data locality can potentially be exploited in two dimensions from a given processor's viewpoint. Because of this, maximizing the number of communications with any of four neighbors (instead of other nodes) helps improve performance. Similar examples can be given from the areas of embedded sensor processing and 3D systems as well. In this application domain, we make two specific contributions. First, we show how array data of a loop-intensive application can be mapped onto a two-dimensional mesh so that the communication distances between the nodes are reduced. Second, we discuss how code restructuring through loop transformation can help us achieve better data locality in the two-dimensional space.

Original languageEnglish (US)
Title of host publicationProceedings of theICCAD-2005
Subtitle of host publicationInternational Conference on Computer-Aided Design
Pages275-278
Number of pages4
DOIs
StatePublished - Dec 1 2005
EventICCAD-2005: IEEE/ACM International Conference on Computer-Aided Design, 2005 - San Jose, CA, United States
Duration: Nov 6 2005Nov 10 2005

Publication series

NameIEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD
Volume2005
ISSN (Print)1092-3152

Other

OtherICCAD-2005: IEEE/ACM International Conference on Computer-Aided Design, 2005
CountryUnited States
CitySan Jose, CA
Period11/6/0511/10/05

Fingerprint

Communication
Sensors
Processing

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design

Cite this

Kandemir, M. (2005). Two-dimensional data locality: Definition, abstraction, and application. In Proceedings of theICCAD-2005: International Conference on Computer-Aided Design (pp. 275-278). [1560077] (IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD; Vol. 2005). https://doi.org/10.1109/ICCAD.2005.1560077
Kandemir, Mahmut. / Two-dimensional data locality : Definition, abstraction, and application. Proceedings of theICCAD-2005: International Conference on Computer-Aided Design. 2005. pp. 275-278 (IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD).
@inproceedings{687c058c64344224b5f49d89e3752503,
title = "Two-dimensional data locality: Definition, abstraction, and application",
abstract = "Data locality has been a central theme in the compiler optimization world for a long time. Most of the prior compiler techniques try to optimize data locality in a one-dimensional linear address space. However, there are many problems out there where the domain for data locality can be two or higher dimensional. For example, in a two-dimensional mesh network environment, each node has connections with its four neighbors, and therefore, the data locality can potentially be exploited in two dimensions from a given processor's viewpoint. Because of this, maximizing the number of communications with any of four neighbors (instead of other nodes) helps improve performance. Similar examples can be given from the areas of embedded sensor processing and 3D systems as well. In this application domain, we make two specific contributions. First, we show how array data of a loop-intensive application can be mapped onto a two-dimensional mesh so that the communication distances between the nodes are reduced. Second, we discuss how code restructuring through loop transformation can help us achieve better data locality in the two-dimensional space.",
author = "Mahmut Kandemir",
year = "2005",
month = "12",
day = "1",
doi = "10.1109/ICCAD.2005.1560077",
language = "English (US)",
isbn = "078039254X",
series = "IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD",
pages = "275--278",
booktitle = "Proceedings of theICCAD-2005",

}

Kandemir, M 2005, Two-dimensional data locality: Definition, abstraction, and application. in Proceedings of theICCAD-2005: International Conference on Computer-Aided Design., 1560077, IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD, vol. 2005, pp. 275-278, ICCAD-2005: IEEE/ACM International Conference on Computer-Aided Design, 2005, San Jose, CA, United States, 11/6/05. https://doi.org/10.1109/ICCAD.2005.1560077

Two-dimensional data locality : Definition, abstraction, and application. / Kandemir, Mahmut.

Proceedings of theICCAD-2005: International Conference on Computer-Aided Design. 2005. p. 275-278 1560077 (IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD; Vol. 2005).

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

TY - GEN

T1 - Two-dimensional data locality

T2 - Definition, abstraction, and application

AU - Kandemir, Mahmut

PY - 2005/12/1

Y1 - 2005/12/1

N2 - Data locality has been a central theme in the compiler optimization world for a long time. Most of the prior compiler techniques try to optimize data locality in a one-dimensional linear address space. However, there are many problems out there where the domain for data locality can be two or higher dimensional. For example, in a two-dimensional mesh network environment, each node has connections with its four neighbors, and therefore, the data locality can potentially be exploited in two dimensions from a given processor's viewpoint. Because of this, maximizing the number of communications with any of four neighbors (instead of other nodes) helps improve performance. Similar examples can be given from the areas of embedded sensor processing and 3D systems as well. In this application domain, we make two specific contributions. First, we show how array data of a loop-intensive application can be mapped onto a two-dimensional mesh so that the communication distances between the nodes are reduced. Second, we discuss how code restructuring through loop transformation can help us achieve better data locality in the two-dimensional space.

AB - Data locality has been a central theme in the compiler optimization world for a long time. Most of the prior compiler techniques try to optimize data locality in a one-dimensional linear address space. However, there are many problems out there where the domain for data locality can be two or higher dimensional. For example, in a two-dimensional mesh network environment, each node has connections with its four neighbors, and therefore, the data locality can potentially be exploited in two dimensions from a given processor's viewpoint. Because of this, maximizing the number of communications with any of four neighbors (instead of other nodes) helps improve performance. Similar examples can be given from the areas of embedded sensor processing and 3D systems as well. In this application domain, we make two specific contributions. First, we show how array data of a loop-intensive application can be mapped onto a two-dimensional mesh so that the communication distances between the nodes are reduced. Second, we discuss how code restructuring through loop transformation can help us achieve better data locality in the two-dimensional space.

UR - http://www.scopus.com/inward/record.url?scp=33751426912&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33751426912&partnerID=8YFLogxK

U2 - 10.1109/ICCAD.2005.1560077

DO - 10.1109/ICCAD.2005.1560077

M3 - Conference contribution

AN - SCOPUS:33751426912

SN - 078039254X

SN - 9780780392540

T3 - IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD

SP - 275

EP - 278

BT - Proceedings of theICCAD-2005

ER -

Kandemir M. Two-dimensional data locality: Definition, abstraction, and application. In Proceedings of theICCAD-2005: International Conference on Computer-Aided Design. 2005. p. 275-278. 1560077. (IEEE/ACM International Conference on Computer-Aided Design, Digest of Technical Papers, ICCAD). https://doi.org/10.1109/ICCAD.2005.1560077