Alignment and distribution is NOT (always) NP-hard

Vincent Boudet 1 Fabrice Rastello 1 Yves Robert 1
1 REMAP - Regularity and massive parallel computing
Inria Grenoble - Rhône-Alpes, LIP - Laboratoire de l'Informatique du Parallélisme
Abstract : An efficient algorithm to simultaneously implement array alignment and data/computation distribution is introduced and evaluated. We re-visit previous work of Li and Chen (J. Li and M. Chen, 1990; 1991), and we show that their alignment step should not be conducted without preserving the potential parallelism. In other words, the optimal alignment may well sequentialize computations, whatever the distribution afterwards. We provide an efficient algorithm that handles alignment and data/computation distribution simultaneously. The good news is that several important instances of the whole alignment/distribution problem have polynomial complexity, while alignment itself is NP-complete (J. Li and M. Chen, 1990)
Complete list of metadatas

https://hal.inria.fr/hal-00856851
Contributor : Equipe Roma <>
Submitted on : Monday, September 2, 2013 - 4:04:54 PM
Last modification on : Friday, November 23, 2018 - 1:40:03 PM

Links full text

Identifiers

Collections

Citation

Vincent Boudet, Fabrice Rastello, Yves Robert. Alignment and distribution is NOT (always) NP-hard. ICPADS'98, Taiwan, Dec 1998, Taiwan, China. pp.648-657, ⟨10.1109/ICPADS.1998.741148⟩. ⟨hal-00856851⟩

Share

Metrics

Record views

300