Thomas Gerhold

MEGAFLOW: Parallel complete aircraft CFD

Abstract As a consequence of the worldwide competition in the aircraft market requirements for the accurate prediction of aerodynamic performance and optimization of configurations have increased very much. More sophisticated wind tunnel as well as high quality CFD techniques have become necessary and essential tools for aircraft industry aerodynamic development groups. This requires an ongoing struggle for efficiency improvements where parallel computing is one of the major issues. This paper considers the MEGAFLOW activities in the area of parallel flow solvers including their application and use in the industrial framework. It describes the parallelization principles for the structured multi-block Navier–Stokes code FLOWer and the unstructured Navier–Stokes code TAU. Principle results for industrial applications are given with respect to efficiency, speed-up, load balancing, etc. Computational examples demonstrate the quality of the solvers for 3D flow over wing/body configurations as well as complete transport aircraft.

GASPI – A Partitioned Global Address Space Programming Interface

At the threshold to exascale computing, limitations of the MPI programming model become more and more pronounced. HPC programmers have to design codes that can run and scale on systems with hundreds of thousands of cores. Setting up accordingly many communication buffers, point-to-point communication links, and using bulk-synchronous communication phases is contradicting scalability in these dimensions. Moreover, the reliability of upcoming systems will worsen.

HICFD: Highly Efficient Implementation of CFD Codes for HPC Many-Core Architectures

The objective of the German BMBF research project Highly Efficient Implementation of CFD Codes for HPC Many-Core Architectures (HICFD) is to develop new methods and tools for the analysis and optimization of the performance of parallel computational fluid dynamics (CFD) codes on high performance computer systems with many-core processors. In the work packages of the project it is investigated how the performance of parallel CFD codes written in C can be increased by the optimal use of all parallelism levels. On the highest level Message Passing Interface (MPI) is utilized. Furthermore, on the level of the many-core architecture, highly scaling, hybrid OpenMP/MPI methods are implemented. On the level of the processor cores the parallel Single Instruction Multiple Data (SIMD) units provided by modern CPUs are exploited.