Parallel performance for a real time Lattice Boltzmann code

This paper details a real time Lattice Boltzmann solver for computing unsteady wakes. The formulation of the lattice Boltzmann method is first presented followed by a discussion of boundary conditions. This is followed by the details of how the basic algorithm was improved to increase single-core efficiency. A discussion of high Reynolds number implementations, and their effect on the overall computational cost follows, and examples are presented alongside performance data on a variety of CPUs.