- New paper submitted: FLEXI: A high order discontinuous Galerkin framework for hyperbolic-parabolic conservation laws October 25, 2019
High order (HO) schemes are attractive candidates for the numerical solution of multiscale problems occurring in fluid dynamics and related disciplines. Among the HO discretization variants, discontinuous Galerkin schemes offer a collection of advantageous features which have lead to a strong increase in interest in them and related formulations in the last decade. The methods have matured sufficiently to be of practical use for a range of problems, for example in direct numerical and large eddy simulation of turbulence. However, in order to take full advantage of the potential benefits of these methods, all steps in the simulation chain must be designed and executed with HO in mind. Especially in this area, many commercially available closed-source solutions fall short. In this work, we therefor present the FLEXI framework, a HO consistent, open-source simulation tool chain for solving the compressible Navier-Stokes equations in a high performance computing setting. We describe the numerical algorithms and implementation details and give an overview of the features and capabilities of all parts of the framework. Beyond these technical details, we also discuss the important, but often overlooked issues of code stability, reproducibility and user-friendliness. The benefits gained by developing an open-source framework are discussed, with a particular focus on usability for the open-source community. We close with sample applications that demonstrate the wide range of use cases and the expandability of FLEXI and an overview of current and future developments.
Link: Paper on arXiv
- Workshop on Efficiency in Computational Science, Cologne, Sep 25th, 2019 September 17, 2019
On Wednesday, September 25th, 2019, the Workshop on Efficiency in Computational Science will take place at the Department of Mathematics and Computer Science. This workshop brings together researchers in the field of computational science to share and discuss their work where it relates to efficiency. Here, efficiency is understood in a very broad sense, including numerical method development, serial and parallel algorithms, implementation, and hardware aspects. The intention is to provide an informal environment that encourages the exchange of novel ideas and untested approaches. Therefore, the speakers are asked to put an emphasis on work in progress and unsolved issues, and to not restrict themselves to sharing “camera-ready” results only. Ultimately, the goal is to get a fresh perspective on common challenges, to establish new connections across institutional and discipline boundaries, and to identify potential for future scientific collaborations.
13:10 Efficiency challenges in adaptive parallel multiphysics simulations
13:40 Current HPC developments in the TRACE flow solver
14:10 Towards Large Scale Continual Learning on Modular High Performance Computers
14:40 Coffee break
15:20 Vectorization of high-order DG and adaptive linearization
15:50 Promises and Challenges of Dispersion Relation Preserving Finite Difference Methods
16:20 Structural modelling for helicopter simulation – or: making small problems even smaller
16:50 Coffee, discussions & open end
- Viktor Linders from Lund University to visit NumSim group for two weeks September 12, 2019
From September 16th through September 27th, Dr. Viktor Linders from Lund University is going to visit the NumSim group by invitation of Dr. Michael Schlottke-Lakemper. Together with researchers from RWTH Aachen University, they are working on high-order methods with optimized dispersive properties for aeroacoustics simulations. During his stay in Cologne, Dr. Linders will share some of his previous results on summation-by-parts methods in a talk at the Workshop on Efficiency in Computational Science.