Enhancing spatial locality via data layout optimizations

M. Kandemir, A. Choudhary, J. Ramanujam, N. Shenoy, P. Banerjee

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

11 Scopus citations

Abstract

This paper aims to improve locality of references by suitably choosing array layouts. We use a new definition of spatial reuse vectors that takes into account memory layout of arrays. This capability creates two opportunities. First, it allows us to develop an array restructuring framework based on a combination of hyperplane theory and reuse vectors. Second, it allows us to observe the effect of different array layout optimizations on spatial reuse vectors. Since the iteration space based locality optimizations also change the spatial reuse vectors, our approach allows us to compare the iteration-space based and data-space based approaches in terms of their effects on spatial reuse vectors. We illustrate the effectiveness of our technique using an example from the BLAS library on the SGI Origin distributed shared-memory machine.

Original languageEnglish (US)
Title of host publicationEuro-Par 1998 Parallel Processing - 4th International Conference, Proceedings
PublisherSpringer Verlag
Pages422-434
Number of pages13
ISBN (Print)3540649522, 9783540649526
DOIs
StatePublished - 1998
Event4th International Conference on Parallel Processing, Euro-Par 1998 - Southampton, United Kingdom
Duration: Sep 1 1998Sep 4 1998

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1470 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other4th International Conference on Parallel Processing, Euro-Par 1998
Country/TerritoryUnited Kingdom
CitySouthampton
Period9/1/989/4/98

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Enhancing spatial locality via data layout optimizations'. Together they form a unique fingerprint.

Cite this