Complex fluid simulations with the parallel tree-based Lattice Boltzmann solver Musubi

• Published in: Journal of computational science 2014-09, Vol.5 (5), p.784-794
• Main Authors: Hasert, Manuel, Masilamani, Kannan, Zimny, Simon, Klimach, Harald, Qi, Jiaxing, Bernsdorf, Jörg, Roller, Sabine
• Format: Article
• Language: English
• Subjects: Computational fluid dynamics; Lattice Boltzmann Method; Octree; Parallel computing
• ISSN: 1877-7503; 1877-7511
• DOI: 10.1016/j.jocs.2013.11.001

•We present the open source Lattice Boltzmann solver Musubi for massively parallel computing systems.•It allows arbitrarily complex simulation domains and user-defined compute stencils in a modularized software design.•Musubi integrates into the end-to-end parallel simulation framework APES.•It is designed for a high sustained computational performance and achieves good serial performance by vectorizing kernels.•We present the high computational performance on two high-ranked Top500 supercomputing systems. We present the open source Lattice Boltzmann solver Musubi. It is part of the parallel simulation framework APES, which utilizes octrees to represent sparse meshes and provides tools from automatic mesh generation to post-processing. The octree mesh representation enables the handling of arbitrarily complex simulation domains, even on massively parallel systems. Local grid refinement is implemented by several interpolation schemes in Musubi. Various kernels provide different physical models based on stream-collide algorithms. These models can be computed concurrently and can be coupled with each other. This paper explains our approach to provide a flexible yet scalable simulation environment and elaborates its design principles and implementation details. The efficiency of our approach is demonstrated with a performance evaluation on two supercomputers and a comparison to the widely used Lattice Boltzmann solver Palabos.