[PoloLaccro]

Hello,

I have a problem to parallelize my case. I am using waveFoam solver to study the motion of a rectangular buoy placed in front of a vertical wall, this buoy is constrain to heaving motion only. I am using OpenFOAM v1912.

I am using a dynamic multi-block mesh, in which the block where the buoy is located is between two areas delimited by AMI patches, shown in red on my mesh.

When I run on a single processor the case runs fine.

However when I try to parallelize my case, crashes occur depending the mesh refinement, the number of processor or the parallelization method.

When all the moving part is on a single processor the case seems runs fine, but when the moving part is cut in different processor, crashes occur when the amplitude of the float becomes large (when the wave impact the buoy).

I would have liked to know if you have any idea where these problems come from, if they come from that the parallelization via waveFoam or parallelization of cases with AMI patch ? Do you have an idea to fix it ? I need to do this test in 3D and whitout paralellization is very long. I remain at your disposal for more information.

I attach to this mail my test case and an image of the mesh.
The parameters are written in the prepareCase.py file.

Thank you !
Have a nice day,

Paul