A study on parallelizing grid-based fluid simulations on the CPU
Sandip Pawar, Parag Chaudhuri · 2014
We present a study based on our initial efforts to parallelize a grid-based fluid solver. The standard Eulerian grid-based fluid solver discretizes the Navier Stokes equation on a regular grid and computes the fluid velocity and pressure in each grid cell. We use OpenMP to parallelize this computation and study the behaviour of our implementation on various machines with different number of CPU cores. In this study, we do not use the GPU to accelerate our simulation. We concentrate on understanding the behaviour of the simulation on the CPU instead.