Researcher - Adaptive mesh refinement techinques exploting DRL (R3) - AI4S per el Barcelona Supercomputing Center

There is a need to increase the capabilities of current Computational Fluid Dynamics tools for engineering design by re-engineering them for extreme-scale parallel computing platforms. The backbone of the Large-scale Computational Fluid Dynamics (LS/CFD) team is centred on the fact that today, the capabilities of leading-edge emerging HPC architectures are not fully exploited by industrial simulation tools. Current state-of-the-art industrial solvers do not take sufficient advantage of the immense capabilities of new hardware architectures, such as streaming processors or many-core platforms. A combined research effort focusing on novel numerical methods, more accurate physical models, algorithms and HPC application is the only way to make it possible to develop and advance simulation tools to meet the needs of the European industry. The LS/CFD team will focus on the development of numerical tools, turbulence models, multi-physics algorithms, data-driven methodologies and large-scale industrial simulations.

The selected candidate will lead activities in the LS/CFD group adaptive mesh refinement techniques for high order finite elements. Required components of such a project include:

• General aspects adaptive mesh refinement techniques for high order methods
• Machine-learning techniques to optimize the mesh refinement procedure (using DRL and ML surrogates)
• Exascale HPC simulations in complex domains with adaptive mesh
• High-performance on modern node architectures

Successful development of such a project would have translational benefits over a broad range of applications of national and international. Its success requires a researcher who understands the full stack outlined above.

Key Duties

• Participate GPU activities to be carried out by BSC in the LS/CFD team
• Interact with the different collaborators of the LS/CFD team to carry out our collaborative research
• Research on GPU porting of dynamic adaptive mesh refinement techniques of CFD codes.
• Contribute in scientific papers publications
• Research on ML integration for CFD codes for mesh adaptive refinement.
• Lead the meshing activities in the LS/CFD group

Data de tancament: Divendres, 01 Novembre, 2024

Més informació

Més posts de Recerca