Compiler Optimizations for Improving Data Locality

Compiler Optimizations for Improving Data Locality (1994)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Steve Carr , Kathryn S. McKinley , Chau-Wen Tseng

Venue:	IN PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE ON ARCHITECTURAL SUPPORT FOR PROGRAMMING LANGUAGES AND OPERATING SYSTEMS
Citations:	219 - 17 self

BibTeX

@INPROCEEDINGS{Carr94compileroptimizations,
    author = {Steve Carr and Kathryn S. McKinley and Chau-Wen Tseng},
    title = {Compiler Optimizations for Improving Data Locality},
    booktitle = {IN PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE ON ARCHITECTURAL SUPPORT FOR PROGRAMMING LANGUAGES AND OPERATING SYSTEMS},
    year = {1994},
    pages = {252--262},
    publisher = {}
}

OpenURL

Abstract

In the past decade, processor speed has become significantly faster than memory speed. Small, fast cache memories are designed to overcome this discrepancy, but they are only effective when programs exhibit data locality. In this paper, we present compiler optimizations to improve data locality basedon a simple yet accurate cost model. The model computes both temporal and spatial reuse of cache lines to find desirable loop organizations. The cost model drives the application of compound transformations consisting of loop permutation, loop fusion, loop distribution, and loop reversal. We demonstrate that these program transformations are useful for optimizing many programs. To validate our optimization strategy, we implemented our algorithms and ran experiments on a large collection of scientific programs and kernels. Experiments with kernels illustrate that our model and algorithm can select and achieve the best performance. For over thirty complete applications, we executed the origi...

Keyphrases

compiler optimization improving data locality data locality compound transformation cache memory loop fusion loop reversal processor speed loop distribution optimization strategy cache line spatial reuse scientific program cost model desirable loop organization loop permutation accurate cost model program transformation large collection thirty complete application memory speed many program past decade