Intel® Parallel Computing Center at Computational Fluid Dynamic department (ONERA)

O N E R A logo

Principal Investigators:

Alain Refloch

Alain Refloch integrated ONERA in 1990 and was in charge of the user’s support for the scientific computation. He was at the origin of the unit ‘Software engineering and HPC’ in 2000. He became project Leader of the CEDRE software in 2003 (see reference paper CEDRE Software) and joined the Computational Fluid Dynamics and AeroAcoustics Department. A. Refloch is a member of the scientific council of the ORAP since 2009.

He was the co-organizer of the International Workshop on High Performance Computing – Computational Fluid Dynamics (HPC-CFD) in Energy/Transport Domains (16th IEEE High Performance Computing and Communication in Paris and ISC'15 in Frankfurt). Today he’s Special Advisor for HPC.

Ivan Mary

Ivan Mary obtained his PhD in 1999 at Paris-Orsay University in the field of numerical methods for CFD. He joined ONERA in 2000 with the mission to develop methods and software allowing efficient Large Eddy Simulation (LES) and Direct Numerical (DNS) simulations of turbulent flow around complex configurations. He has supervised around ten PhD students during the last fifteen years in the fields of numerical method, fluid dynamics and turbulence modelling. Since 2011, he has focused a large part of his work on HPC, because this is a crucial point for unsteady computations of turbulent flows based (re-engineering, coarse-grain OpenMP parallelization, and vectorization). Since 2015, He is in charge of the Fast demonstrator, which must provide the HPC basis of the next generation elsAsoftware.


The ONERA CFD department develops and supports fluid dynamics software for decades both for its own research and for industrial partners in the aeronautical domain. Nowadays the elsA software, developed at ONERA since 1997, is one of the major CFD tools used by Airbus, Eurocopter and Safran. In their design services, it is massively employed to optimize airplane performance (noise or energy consumption reduction, safety improvement). Due to environmental constraint, noise reduction in the vicinity of airports has become a major challenge for aircraft manufacturer. The noise radiated during the landing phase is due to turbulent vortices generated by landing gears and flaps in the wings, which act like powerful whistles. The numerical simulation of the generated noise requires to handle the complex detailed geometry of landing gear or flaps and to solve billions of unknowns at each time step to describe the time evolution of turbulence vortices during millions of time step in order to compute few seconds of the physical time.

Therefore HPC capabilities, complex geometries (re) meshing and multiphysics coupling (noise generator and propagator) are crucial points for the efficiency of the software to obtain a solution in a reasonable time. For these reasons, a demonstrator named FAST (Flexible Aerodynamic Solver Technology) is under development since one to two year in order to prepare a major evolution of elsA in the coming years. This demonstrator aims to provide a software architecture and numerical techniques which will allow better flexibility, evolutivity and efficiency in order to perform simulations out of reach with the actual CFD tools. Thanks to previous expertise, services reclaimed by CFD simulations (pre/post-processing, boundary conditions, solvers, coupling, etc.) are provided by different Python modules in FAST, whereas the CFD General Notation System (CGNS) standard is adopted as a data model, but also for the implementation of this data model in order to facilitate interoperability between modules. To improve flexibility in the meshing of complex geometrical details, an automatic cartesian grid generator, immersed boundary condition and chimera technique will be employed during the present Intel® Parallel Computing Center(s) (Intel® PCC) project to compute the noise generated by the LAGOON landing gear configuration (Lagoon). Thanks to code modernization (memory access, vectorization, etc.) we aim to reduce by at least one order of magnitude the CPU cost of this kind of computation on actual Intel® Xeon® and future Intel® Xeon Phi™ processor family.

Complex Structures of Dynamic Stall by LES
Complex Structures of Dynamic Stall by LES by Ivan Mary (ONERA)


Related Websites:

For more complete information about compiler optimizations, see our Optimization Notice.