DETECTION OF SCANS

Most automatic parallelizes are based on the detection of independent operations. Dependence analysis is mainly a syntactical process, in which the actual data transformations are ignored. There is another source of parallelism, which relies on semantical information, namely the detection of reducti...

Full description

Saved in:
Bibliographic Details
Published inParallel algorithms and applications Vol. 15; no. 3-4; pp. 229 - 263
Main Authors REDON, XAVIER, FEAUTRIER, PAUL
Format Journal Article
LanguageEnglish
Published Taylor & Francis Group 01.12.2000
Subjects
Automatic parallelization
D.2.4 Software/Program Verification (Formal methods)
D.3.4 Processors (Optimization)
F.2.2 Nonnumencal Algorithms and Problem (Pattern matching)
F.3.2 Semantics of Programming Languages (Program analysis)
Programs normalization
programs transformations
Reductions
Scans
Online AccessGet full text
ISSN1063-7192
DOI10.1080/01495730008947357

Cover

More Information
Summary:Most automatic parallelizes are based on the detection of independent operations. Dependence analysis is mainly a syntactical process, in which the actual data transformations are ignored. There is another source of parallelism, which relies on semantical information, namely the detection of reductions and scans. Scans and reductions are quite frequent in scientific codes and are implemented efficiently on most parallel computers. We present here a new Scan detector which is based on the normalization of systems of recurrence equations. This allows the detection of scans in loops nests of arbitrary depth and on multi-dimensional arrays, and gives a uniform treatment for scalar reductions, array reductions, and arrays of reductions.
ISSN:1063-7192
DOI:10.1080/01495730008947357