A Novel A.I Enhanced Reservoir Characterization with a Combined Mixture of Experts -- NVIDIA Modulus based Physics Informed Neural Operator Forward Model

Autor: Etienam, Clement, Juntao, Yang, Said, Issam, Ovcharenko, Oleg, Tangsali, Kaustubh, Dimitrov, Pavel, Hester, Ken

We have developed an advanced workflow for reservoir characterization, effectively addressing the challenges of reservoir history matching through a novel approach. This method integrates a Physics Informed Neural Operator (PINO) as a forward model w

Externí odkaz: http://arxiv.org/abs/2404.14447

Apports des architectures hybrides à l'imagerie profondeur : étude comparative entre CPU, APU et GPU.

Autor: Said, Issam

Les compagnies pétrolières s'appuient sur le HPC pour accélérer les algorithmes d'imagerie profondeur. Les grappes de CPU et les accélérateurs matériels sont largement adoptés par l'industrie. Les processeurs graphiques (GPU), avec une grande

Externí odkaz: http://www.theses.fr/2015PA066531/document

Exploiting temporal data reuse and asynchrony in the reverse time migration.

Autor: Qu, Long, Abdelkhalak, Rached, Ltaief, Hatem, Said, Issam, Keyes, David

Publikováno v: International Journal of High Performance Computing Applications; Mar2023, Vol. 37 Issue 2, p132-150, 19p

Estimation de la reproductibilité numérique dans les environnements hybrides CPU-GPU

Autor: Jézéquel, Fabienne, Lamotte, Jean-Luc, Said, Issam

Publikováno v: CANUM, mini-symposium ``Recherche reproductible''
CANUM (Congrès National d'Analyse Numérique)
CANUM (Congrès National d'Analyse Numérique), May 2016, Obernai, France

National audience

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::88e8dd34e1b4d9c77c84c70494e286fb
https://hal.science/hal-01358094

Hybrid strategy for stencil computations on the APU

Autor: Eberhart, Pacôme, Said, Issam, Fortin, Pierre, Calandra, Henri

Publikováno v: Proceedings of the 1st International Workshop on High-Performance Stencil Computations
The 1st International Workshop on High-Performance Stencil Computations
The 1st International Workshop on High-Performance Stencil Computations, Jan 2014, Vienna, Austria. pp.43-49

International audience; Stencil computations are very regular and well adapted to GPU execution. However, the PCI-E bus that connects a discrete GPU to the system memory has a relatively low bandwidth when compared to the GPU compute power. The AMD A

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=dedup_wf_001::209a05960e8bff9e81cec762bab26797
https://hal.archives-ouvertes.fr/hal-01213556