[qjh888]

Hi Foamers,

I try to run the case with mpi parallel.
However, I've faced with some problems.

1. When running in parallel, the case is diverged. But when running in single core, it is converged sufficiently.
2. when running in parallel, the case is quite slow rather than running in single core.
3. After using decomposePar utility, I cannot open the mesh in paraview for processorN. The paraview offers the error message like that:
   

     p, li { white-space: pre-wrap; }  4725
    vtkOpenFOAMReaderPrivate (0x9e7ff88): Numbers of faces in faces 346703 and owner 343167 does not match
    
   
    
   
    ERROR: In /build/buildd/paraview-4.0.1/VTK/Common/ExecutionModel/vtkExecutive.cxx, line 754
    vtkPVCompositeDataPipeline (0xa6ed908): Algorithm vtkPOpenFOAMReader(0xa630578) returned failure for request: vtkInformation (0xad3a448)
      Debug: Off
      Modified Time: 119955
      Reference Count: 1
      Registered Events: (none)
      Request: REQUEST_DATA
      ALGORITHM_AFTER_FORWARD: 1
      FROM_OUTPUT_PORT: 0
      FORWARD_DIRECTION: 0

Could anyone help me out of the situation?
Thanks!
Janry