This position will work as a post-doctoral researcher in the European project NextSim. The backbone of NextSim is centred on the fact that, today, the capabilities of leading-edge emerging HPC architectures are not fully exploited by industrial simulation tools. Given the expertise of the partners, we will focus on aeronautical problems, but this upstream research can be applied to a variety of industries, some of them mentioned above, where the models used for design rely on the numerical discretization and integration of partial differential equations (e.g. automotive, wind energy, etc.). 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. The computing world has changed over recent years, with processor clock speeds no longer increasing, and multicore/heterogeneous platforms now dominating. A combined research effort focusing on algorithms and HPC is the only way to make possible to develop and advance simulation tools in order to meet the needs of the European aeronautical industry.
NextSim will focus on the development of the numerical flow solver CODA. The CODA solver includes classic finite volume capabilities and new highly accurate high order discontinuous Galerkin schemes, all specifically tailored for aeronautical applications, and available to all partners. The CODA solver will be the new reference solver for aerodynamic applications inside AIRBUS group (including aircraft and helicopters), as such, it will be used by AIRBUS group (including Civil, Helicopters, Space, Military) for the design of a variety of engineering applications, having a significant impact in the aeronautical market.
And, to facilitate the transfer of knowledge to the wider simulations community, the developments of CODA in NextSim will be made available through mini-apps (simple feature specific solvers) and the various existing open source solvers of the partners (such as Alya at BSC, Horses3D at UPM or Aghora at ONERA), widening the access to codes and fostering transfer of know-how to user communities.
To turn CODA into the future aerodynamic industrial solver, NextSim defines the following objectives:
1.) Re-engineer industrial aerodynamic solver CODA for the efficient use of parallel heterogeneous HPC-petascale architectures available now.
2.) Improve the convergence of current numerical algorithms for heterogeneous HPC systems that may comprise many-core and GP-GPU. Whatever the numerical approach used (Finite Volume, FV, of High Order Methods, HoM) and the number and complexity of the partial differential equations (PDE) considered: Reynolds Averaged Navier Stokes equations (RANS) or Turbulent Scales Resolving Simulations (TSRS).
3.) Appropriate for adaptive simulations for which the number of DoF evolves during the iterative process while solving the system of PDE, e.g. in the perspective of h/p adaptation. This includes appropriate static and dynamic load balancing strategies
4.) Portable and versatile versus various types of HPC hardware.
5.) Bring the HoM technology to the same level of maturity, robustness and computational efficiency than FV to envisage its deployment in operations for certain class of problem
6.) Overcome barriers to achieving extreme scale computing by choosing the appropriate programming model; replacing synchronous with asynchronous communication; using parallelism at many levels: from the instruction set at the functional unit level, to large grain processes at the multi-node scale; improving energy efficiency.
7.) Solve a series of market aeronautical test cases but currently unfeasible because of their computational costs, and perform a comparison of the “before and after” simulation time and accuracy.
8.) Disseminate the project results through the “mini-Apps” idea, (see details below) or kernels for numerical simulation that will be open source, enabling researchers and firms external to the project to benefit quickly from any innovations.
9.) Integrating the user in the NextSim evolution. NextSim can only be successful and sustainable if the stakeholders in the consortium and potential consumers of the centre’s services are involved, get informed and can provide feedback in an adequate way.