Semi-Automatic Generation of Adaptive Codes

Abstract : Compiler automatic optimization and parallelization techniques are well suited for some classes of simulation or signal processing applications, however they usually don't take into account domain-specific knowledge nor the possibility to change or to remove some computations to achieve " good enough " results. Quite differently, production simulation and signal processing codes have adaptive capabilities: they are designed to compute precise results only where it matters if the complete problem is not tractable or if computation time must be short. In this paper, we present a new way to provide adaptive capabilities to compute-intensive codes automatically. It relies on domain-specific knowledge provided through special pragmas by the programmer in the input code and on polyhedral compilation techniques to continuously regenerate at runtime a code that performs heavy computations only where it matters at every moment. We present a case study on a fluid simulation application where our strategy enables significant computation savings and speedup in the optimized portion of the application while maintaining a good precision, with a minimal effort from the programmer.
Type de document :
Communication dans un congrès
IMPACT 2017 - 7th International Workshop on Polyhedral Compilation Techniques, Jan 2017, Stockholm, Sweden. pp.1-7
Liste complète des métadonnées

Littérature citée [16 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01655456
Contributeur : Cédric Bastoul <>
Soumis le : lundi 4 décembre 2017 - 22:08:24
Dernière modification le : jeudi 29 mars 2018 - 09:10:05

Fichier

impact-17-semi-automatic-gener...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01655456, version 1

Citation

Maxime Schmitt, César Sabater, Cédric Bastoul. Semi-Automatic Generation of Adaptive Codes. IMPACT 2017 - 7th International Workshop on Polyhedral Compilation Techniques, Jan 2017, Stockholm, Sweden. pp.1-7. 〈hal-01655456〉

Partager

Métriques

Consultations de la notice

170

Téléchargements de fichiers

40