[Paraview] Ongoing Work to Release Unused Memory from Pipeline

[Jérôme]

I don't know if it applies to your pipelines, but each time I write an XML
ServerManager to expose a VTK filter in ParaView, I add a ckeckbox for the
vtkAlgorithm's function "ReleaseDataFlag" :

        <IntVectorProperty
                           name="ReleaseDataFlag"
                           command="SetReleaseDataFlag"
                           number_of_elements="1"
                           default_values="0"
                           animateable="0">
            <BooleanDomain name="boolean"/>
            <Documentation>
                  Turn release data flag on or off for all output
ports.
            </Documentation>
         </IntVectorProperty>

Obviously, it implies that the filters with RealeaseDataFlag set to true
will reexecute at each update, even it they are not marked as modified.
You can find such examples in my github repo that exposes VTK native filters
to ParaView: http://github.com/jeromevelut/Peavip . All the
vtkImageAlgorithm derived filter propose this option.

HTH
Jerome

2011/7/8 Berk Geveci <berk.geveci at kitware.com>

> If this is something that you are interested in exploring, we can help
> to some extent.
>
> On Thu, Jul 7, 2011 at 3:22 PM, Utkarsh Ayachit
> <utkarsh.ayachit at kitware.com> wrote:
> > Mike,
> >
> > There are currently no such plans in the near future, at the least.
> > But that's definitely something we should start thinking about more
> > seriously as we start dealing with larger and larger datasets.
> >
> > Utkarsh
> >
> > On Wed, Jul 6, 2011 at 10:12 PM, Michael Jackson
> > <mike.jackson at bluequartz.net> wrote:
> >> Is there any ongoing or upcoming work in ParaView to have some sort of
> option to dispose of unused memory from filters that are higher up in the
> pipeline? We are thinking of basing our Materials Science based software
> package on VTK and then make the filters available in ParaView but our data
> sets are already pushing the memory limits of "PCs". Having multiple copies
> of the data in memory would just make a bad situation worse for us. I know
> if I write my own pipeline in VTK I can manually dispose of memory from each
> pipeline step. Just wandering if this type of capability is going to be
> exposed in someway in an upcoming version of ParaView.
> >>
> >> Thanks
> >> ___________________________________________________________
> >> Mike Jackson                      www.bluequartz.net
> >> Principal Software Engineer       mike.jackson at bluequartz.net
> >> BlueQuartz Software               Dayton, Ohio
> >>
> >> _______________________________________________
> >> 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 ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
> >>
> >> Follow this link to subscribe/unsubscribe:
> >> http://www.paraview.org/mailman/listinfo/paraview
> >>
> > _______________________________________________
> > 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 ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
> >
> > Follow this link to subscribe/unsubscribe:
> > http://www.paraview.org/mailman/listinfo/paraview
> >
> _______________________________________________
> 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 ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20110708/40b08fbb/attachment-0001.htm>