Post-Image

Benoît LELANDAIS

Post-Image Post-Image Post-Image

Benoît Lelandais holds an MS of Université Paris-Saclay in computer science. It has held several positions in digital services companies, in IFP Energies Nouvelles and at the French Atomic Energy Commission (CEA) in the field of software design and implementation. He is currently working in a team in charge of providing meshing methods and tools for HPC numerical simulation codes.

After numerous works in model-driven engineering, including the development of DSLs for numerical simulation and their software environments to generate efficient code in an HPC context, his main mission today concerns Magix3D, a 3D meshing tool.

Monilogging for executable domain-specific languages
Dorian Leroy   Benoı̂t Lelandais   Marie-Pierre Oudot   Benoit Combemale  
Proceedings of the 14th ACM SIGPLAN International Conference on Software Language Engineering, p. 2-15, 2021

Applying model-driven engineering to high-performance computing: Experience report, lessons learned, and remaining challenges
Benoı̂t Lelandais   Marie-Pierre Oudot   Benoit Combemale  
Journal of Computer Languages, Elsevier, p. 100919, 2019

Fostering metamodels and grammars within a dedicated environment for HPC: the NabLab environment (tool demo)
Benoı̂t Lelandais   Marie-Pierre Oudot   Benoit Combemale  
Proceedings of the 11th ACM SIGPLAN International Conference on Software Language Engineering, p. 200-204, 2018

Modane: a design support tool for numerical simulation codes
Benoı̂t Lelandais   Marie-Pierre Oudot  
Oil & Gas Science and Technology--Revue d'IFP Energies nouvelles, EDP Sciences, p. 57, 2016

The Arcane Development Framework
Gilles Grospellier   Benoit Lelandais  
Proceedings of the 8th Workshop on Parallel/High-Performance Object-Oriented Scientific Computing, Association for Computing Machinery, 2009

abstract
Abstract

In this paper, we introduce the Arcane software development framework for 2D and 3D numerical simulation codes. First, we describe the Arcane core, the mesh management and the parallelism strategy. Then, we focus on the concepts introduced to speed up the development of numerical codes: numerical modules, variables, entry points and services. We explain the execution model and enumerate the available debugging tools. Finally, the main functionalities of Arcane are described through an example. As a conclusion, we present the future works.