Multi-GPU Implementation of a Hybrid Thermal Lattice Boltzmann Solver using the TheLMA Framework

C. Obrecht (1) , F. Kuznik (1) , Bernard Tourancheau (1) , J.-J. Roux (1)

1 CETHIL - Centre de Thermique de Lyon

In this contribution, a single-node multi-GPU thermal lattice Boltzmann solver is presented. We implement a simplified version of the hybrid model developed by Lallemand and Luo in 2003, which combines multiple-relaxation-time lattice Boltzmann for the fluid flow with a finite-difference method for temperature. The program is based on the TheLMA framework which was developed for that purpose. The chosen implementation and optimisation strategies are described, both for inter-GPU communication and for coupling with the thermal component of the model. Validation and performance results are provided as well.

Mots clés: Thermal lattice Boltzmann method; GPU computing; CUDA

Domaines: Thermique [physics.class-ph]; Thermique [physics.class-ph]

C. Obrecht, F. Kuznik, Bernard Tourancheau, J.-J. Roux. Multi-GPU Implementation of a Hybrid Thermal Lattice Boltzmann Solver using the TheLMA Framework. Computers and Fluids, 2013, 80, pp.269-275. ⟨10.1016/j.compfluid.2012.02.014⟩. ⟨hal-00731152⟩