pimpleDyMFoam on multiple nodes
 

Hello,

I am experiencing some trouble trying to use the solver pimpleDyMFoam in parallel on multiple nodes.
In this case, the computation is slowed (8 CPUs on 2 nodes is twice slower than 8 CPUs on 1 node).
I use scotch to decompose the domain (around 39000 cells by CPU) and GAMG to solve the linear systems.

Do you have any idea about what is the problem?
I would greatly appreciate your help.

AMD