Skip to Main content Skip to Navigation
Reports

An exact algorithm for the constraint satisfaction problem : application to dependance computing in automatic parallelization

Abstract : The constraint satisfaction problem - denoted by CSP - consists in proving the emptiness of a domain defined by a set of constraints or the existence of a solution. Numerous applications arise in the computer science field (artificial intelligence, vectorization, verification of programs,...). In the case of the study of dependence computing in automatic parallelization, classical methods in literature may break down for some instances of CSP, even with small sizes. By constrast the new method - denoted by FAS3T (Fast Algorithm for the Small Size constraints Satisfaction problem Type) - we propose allows an efficient solution of the CSP concrete instances generated by the VATIL vectorizer. Comparative computational results are reported.
Document type :
Reports
Complete list of metadata

https://hal.inria.fr/inria-00075312
Contributor : Rapport de Recherche Inria <>
Submitted on : Wednesday, May 24, 2006 - 5:56:45 PM
Last modification on : Thursday, February 11, 2021 - 2:50:07 PM
Long-term archiving on: : Tuesday, April 12, 2011 - 10:28:27 PM

Identifiers

  • HAL Id : inria-00075312, version 1

Collections

Citation

H. Bennaceur, G. Plateau, F. Thomasset. An exact algorithm for the constraint satisfaction problem : application to dependance computing in automatic parallelization. RR-1246, INRIA. 1990. ⟨inria-00075312⟩

Share

Metrics

Record views

96

Files downloads

56