[syavash]

Dear Foamers,

I have a problem when sampling a plane in parallel mode. In serial mode, everything is fine with sampling data on a plane during run-time. The 2-D sampled planes are properly captured and the flow pattern is reasonable.

However, when I decompose my case and run the job in parallel mode, I get some EXTRA regions outside of the pre-defined sampling space. For instance, if I define a 2-D sampling plane normal to Z-direction, in serial mode it looks fine but in parallel mode some other regions (in different Z-coordinates) are captured along.

I guess that this is a problem with data distribution among the different processes, as I could virtually resolve it by decomposing in a direction normal to the sampling plane (Here in Z-direction). However, this cannot be a real solution as I cannot afford to decompose my other grids in Z-direction.

I wonder if some people have run into this problem before and already have found the solution.

Regards,
Syavash

P.S. I use OF 2.3.1