Load imbalance is an important impediment on the path towards higher degrees of parallelism-especially for engineering codes with their highly unstructured problem domains. In particular, when load conditions change dynamically, efficient mesh partitioning becomes an indispensable ingredient of scalable design. However, popular graph-based methods such as those used by ParMetis require global knowledge, which effectively limits the problem size on distributed-memory machines. On such architectures, space-filling curves (SFCs) offer a memory-efficient alternative and many sophisticated schemes have already been proposed. In this paper, we present a simple strategy based on SFCs that is custom-tailored to the needs of static meshes with dynamically changing computational load. Exploiting the properties of this class of problems, it is not only easy to implement but also reduces memory requirements substantially. Moreover, exclusively relying on MPI collective operations, our load-balancing scheme also offers portable performance across a broad range of HPC systems. Experimental evaluation shows excellent scaling behavior for up to 16,384 cores on a Nehalem-Infiniband system and up to 294,912 processes on a Blue Gene/P system.

No abstract

No abstract

In this paper, we present results on the noise radiated by a supersonic gas jet through a gas injection nozzle which is commonly used in cars powered by compressed natural gas. Direct aeroacoustic simulation results, provided by a discontinuous Galerkin solver for the unsteady compressible Navier Stokes equations, are presented. Due to its local formulation with high order accuracy and low dissipation and dispersion, it is well suited for both flow resolution and wave propagation. The discontinuous Galerkin scheme preserves these properties even on unstructured hybrid grids, allowing to mesh the complex threedimensional geometries. The scheme is fully parallelized and simulations are run on 2048 cores. We show that the sound radiation can be adequately predicted and that sound pressure level spectra are in good agreement with experimental data.