[Paraview] Parallel Data Visualization

[David E DeMarle]

On Fri, Dec 5, 2014 at 10:19 AM, 樱空Sai§ <272586668 at qq.com> wrote:

> Hi all！
>       I connected the client to the pvserver mpi job to test if paraview
> was using all the nodes. I opened Memory inspector, I saw there were many
> processors' information.Then I used the Process Id scalars filter, there
> was only one color. After I used  D3 filter I can see many colors, but
> pvserver occupied much more memory.
>       What can I do to make paraview distribute data to different nodes at
> the time I load the data, without using D3 filter?
>

What file format is this data in? Not all file format readers are able to
split apart their input. Try saving out the results of the D3 filter into
an exodus file, quitting paraview and then starting it up again and loading
the data back in.



>       If I have 4 nodes with 2G memory, is it posssible for me to process
> 6G data?
>

8G total for 6G seems a bit tight, but it _might_ be possible. It depends a
lot on what the data format is and what you want to do with it. Some
operations are more memory thrifty than others. There are a helpful
suggestions and guidelines in the ParaView tutorial and book sections.


> ‍
>
> _______________________________________________
> 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://public.kitware.com/mailman/listinfo/paraview
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20141208/6a4d83a4/attachment.html>