Automatic Code Generation of Distributed Parallel Tasks

Abstract : With the advent of clustered systems, more and more parallel computing is required. However a lot of programming skills is needed to write aparallel codes, especially when you want to benefit from the various parallel architectural resources, with heterogeneous units and complex memory organizations. We present in this paper a method that generates automatically, step by step, a task-parallel distributed code from a sequential program. It has been implemented in an existing source-to-source compiler PIPS. Our approach provides two main advantages 1) all the program transformations are simple and applied on source code, thus are visible by the user, 2) a proof of correctness of the parallelization process can be made. This ensures that we end up with a correct distributed-task program for distributed-memory machines. To our knowledge, it is the first tool that automatically generates a distributed code for task parallelization.
Type de document :
Communication dans un congrès
Liste complète des métadonnées

https://hal-mines-paristech.archives-ouvertes.fr/hal-01431386
Contributeur : Claire Medrala <>
Soumis le : lundi 16 janvier 2017 - 13:47:55
Dernière modification le : lundi 12 novembre 2018 - 10:58:44

Identifiants

  • HAL Id : hal-01431386, version 1

Citation

Nelson Lossing, Corinne Ancourt, François Irigoin. Automatic Code Generation of Distributed Parallel Tasks. Onzième rencontre de la communauté française de compilation, Sep 2016, Aussois, France. ⟨hal-01431386⟩

Partager

Métriques

Consultations de la notice

101

Téléchargements de fichiers

34