Jiaxing Qi / 齐家兴: Efficient Lattice Boltzmann Simulations on Large Scale High Performance Computing Systems, STS (2017), Vol. 3
This thesis presents efforts to attain efficient Lattice Boltzmann simulations on large scale parallel High Performance Computing (HPC) systems. To enable the fluid simulation on non-uniform meshes, a three-dimensional compact interpolation scheme is presented, which greatly reduces the computational effort.
For an efficient utilization of the parallel and distributed computing resources, the key issues in software implementation are discussed. The presented methods are implemented in the Lattice Boltzmann solver – Musubi, and available under a permissive open source license.
Jiaxing Qi (齐家兴) is a scientific researcher in the field of Scientific Computing. He has multiple years‘ experience in developing Computational Fluid Dynamics (CFD) software, which can be run on massively parallel supercomputers with up to thousands of cores. He is one of the major developers of the opensource Lattice Boltzmann solver – Musubi, which has been optimized for various High Performance Computing (HPC) architectures, e.g. Intel x86, IBM PowerPC and NEC SX vector machine.
Jiaxing Qi (齐家兴)
Efficient Lattice Boltzmann Simulations on Large Scale High Performance Computing Systems
Simulation Techniques in Siegen / STS, 2017, Vol. 3
Sabine Roller / Harald Klimach (Ed.)
Siegen: universi 2017, 154 pp.
Preis: 19,- Euro