[vtkusers] Different behavior of vtkXMLPUnstructuredGridWriter in VTK 7.1

[Andy Bauer]

Hi Bill,

I think that this was a bug in the way that VTK was being used but because
of the way that the XMLPUnstructuredGridWriter was implemented, it ended up
not being manifested in VTK pre 7.1. If they tried to do most other
parallel operations (e.g. streamlines, etc.) they would not get the desired
behavior.

If you don't agree with this sentiment, any suggestions on how this would
be fixed? The new behavior is much better for a variety of reasons when
writing out files in parallel when not all MPI processes have data to be
written out and therefore I would think should be the default.

Andy

On Tue, Dec 6, 2016 at 12:13 PM, Bill Lorensen <bill.lorensen at gmail.com>
wrote:

> Andy,
>
> Isn't there a way to handle the issue in a backward compatible way?
> Seems like this user was surprised and a fix to his code was not
> obvious.
>
> Bill
>
>
> On Tue, Dec 6, 2016 at 12:11 PM, Andy Bauer <andy.bauer at kitware.com>
> wrote:
> > Hi,
> >
> > You'll want to enable VTK_Group_MPI (i.e. -DVTK_Group_MPI:BOOL=ON).
> > Technically the "p" in the pvtu writer really corresponds to piece and
> not
> > necessarily parallel. The pieces do map well to SIMD parallelism though.
> >
> > Best,
> > Andy
> >
> > On Tue, Dec 6, 2016 at 12:06 PM, John Peterson <jwpeterson at gmail.com>
> wrote:
> >>
> >>
> >>
> >> On Tue, Dec 6, 2016 at 9:12 AM, Andy Bauer <andy.bauer at kitware.com>
> wrote:
> >>>
> >>> Hi John,
> >>>
> >>> The issue is indeed due to changes that I mentioned before (i.e. only
> >>> data sets that have data get written out and the .pvtu file is adjusted
> >>> accordingly). With this change though interprocess communication is
> required
> >>> which is done the the vtkMultiProcessController::GetGlobalController()
> >>> object which essentially acts as MPI_COMM_WORLD within VTK. You need to
> >>> create a vtkMPIController which derives from
> vtkMultiProcessController. See
> >>> the attached files for how that is done. This should also be backwards
> >>> compatible but if it isn't, please let us know.
> >>
> >>
> >> Thanks for your help.
> >>
> >> It looks like the VTK I built from source apparently does not have MPI
> >> enabled (neither vtkMPIController.h nor libvtkParallelMPI-7.1.dylib is
> >> present in my installation) as it was never required for our code prior
> to
> >> VTK-7.1. Since I'm very much of a cmake novice, do you know what flag I
> >> would pass to cmake when building VTK to turn this on?  My guess is
> >> something like:
> >>
> >> -DModule_vtkParallelMPI:BOOL=ON
> >>
> >> ?
> >>
> >> --
> >> John
> >
> >
> >
> > _______________________________________________
> > Powered by www.kitware.com
> >
> > Visit other Kitware open-source projects at
> > http://www.kitware.com/opensource/opensource.html
> >
> > Please keep messages on-topic and check the VTK FAQ at:
> > http://www.vtk.org/Wiki/VTK_FAQ
> >
> > Search the list archives at: http://markmail.org/search/?q=vtkusers
> >
> > Follow this link to subscribe/unsubscribe:
> > http://public.kitware.com/mailman/listinfo/vtkusers
> >
>
>
>
> --
> Unpaid intern in BillsBasement at noware dot com
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtkusers/attachments/20161206/1f5a7c75/attachment.html>