Executing OpenMP programs on distributed resources using a source-to-source transformation tool

Eric Renault (1, 2, 3, 4) , Charles Ancelin (2) , Oscar Javier Botero Gomez (5) , Willy Ronald Jimenez Freitez (5, 3, 4)

1 R3S-SAMOVAR - Réseaux, Systèmes, Services, Sécurité
2 TSP - RS2M - Département Réseaux et Services Multimédia Mobiles
3 SAMOVAR - Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux
4 CNRS - Centre National de la Recherche Scientifique
5 LOR - Département Logiciels et Réseaux

For the past few years, OpenMP has become the reference to develop applications for shared-memory architectures thanks to an easy way to express parallel opportunities. Therefore, several works have tried to provide an OpenMP compiler for distributed architectures. This article presents how we interfaced the Turing eXtended Language (TXL) with the GCC compiler to implement the Checkpointing Aided Parallel Execution (CAPE) concepts in order to provide an original OpenMP compiler for distributed architectures

Domaines: Calcul parallèle, distribué et partagé [cs.DC]

Eric Renault, Charles Ancelin, Oscar Javier Botero Gomez, Willy Ronald Jimenez Freitez. Executing OpenMP programs on distributed resources using a source-to-source transformation tool. PDCCS 2008 : 21st International Conference on Parallel and Distributed Computing and Communication Systems, Sep 2008, New Orleans, United States. pp.199 - 204. ⟨hal-00447581⟩