[vtkusers] vtkPDataSetWriter Question

[Jeremy Stout]

Thank you for that reference. The files for vtkDistributedDataFilter
do not seem to be included in my distribution of vtk.

Looking over that class, it seems to do exactly what I want it to do.
Unfortunately, I'm having some trouble understanding how the data is
supposed to be arranged among the processes before I apply the filter.
If I am reading the documentation correctly, each process should have
its own chunk of the larger unstructured grid. You then apply the
fiter so you get spatially contiguous unstructured grids. The problem
is that I am reading from a legacy ASCII VTK file, which does not seem
to support reading in "just a piece".  Would it just be possible for
me to read the data file in on the first process and then apply the
filter?

Jeremy Stout

On Thu, 07 Oct 2004 15:42:12 +0200, Jean M. Favre <jfavre at cscs.ch> wrote:
> Jeremy Stout wrote:
> > Hello, I am working with an unstructured grid that contains several
> > hundred thousand cells. I would like to partition this grid and store
> > it in vtk's parallel file format....
> 
> > What I am wondering is whether this class will
> > intelligently partition my grid or if I need to do that myself.
> 
> Check out vtkDistributedDataFilter. This filter redistributes data among
> processors in a parallel application into spatially contiguous
> vtkUnstructuredGrids.
> 
> http://www.vtk.org/doc/nightly/html/classvtkDistributedDataFilter.html
> 
> --
> Dr. Jean M. Favre,                            email:favre @ cscs.ch
> http://www.cscs.ch/about/Favre.php
> CSCS, Swiss National Supercomputing Centre  | Tel:  +41 (91) 610.82.40
> Via Cantonale, 6928 Manno, Switzerland      | Fax:  +41 (91) 610.82.82
> 
>