Skip to content
Daniel Jasiński edited this page Nov 10, 2015 · 6 revisions

RapidCFD can be compiled using standard OpenFOAM build system. So far the build process was tested using gcc 4.8 and CUDA 6.5.

For multi-GPU simulations OpenMPI 1.8.4 compiled with --with-cuda option is required. Following OpenFOAM convention it should be located in ThirdParty-dev directory.

All source files of RapidCFD are compiled using nvcc, therefore the build process takes extremally long compared to the standard build. To enable parallel compilation set WM_NCOMPPROCS environment variable to the number of parallel processes you want to use for the build process.

Clone this wiki locally