Integrating loop and data transformations for global optimization

O'Boyle, MFP

doi:10.1006/jpdc.2001.1815

Integrating loop and data transformations for global optimization

2002-04-01

11

PRE

AI

DOI：10.1006/jpdc.2001.1815

Original

Original request for help

Save

Abstract

En 中文

This paper is concerned with integrating global data transformations and local loop transformations in order to minimize overhead on distributed shared memory machines such as the SGi Origin 2000. By first developing an extended algebraic transformation framework, a new technique to allow the static application of global data transformations, such as partitioning, to reshaped arrays is presented, eliminating the need for expensive temporary copies and hence eliminating any communication and synchronization. In addition, by integrating loop and data transformations, poor spatial locality and expensive array subscripts that may have been introduced can be eliminated. A specific optimization algorithm is derived and applied to well-known benchmarks, where it is shown to give a significant improvement in execution time over existing approaches. (C) 2002 Elsevier Science (USA).

Keywords:

compiler optimization

unified program transformations

rank-modifying transformations

reshaped arrays

auto-parallelization

Integrating loop and data transformations for global optimization

Abstract

Journal

Organization

Cited Papers