Three different implementation of the DLA. The first one is a single core implementatation, the second is an OpenMP implementation and the third is a CUDA implementation
Some renders of the results (low resolution):
rendered with Yocto/GL by Fabio Pellacini (xelatihy on github)