[Paraview] Parallel visualizations of Xdmf datasets

[Jean-Michel Lucas]

Hi everyone,

 

I am trying to use ParaView for parallel visualizations. I have successfully compiled ParaView 4.3.1 and tested it by creating a sphere on several processors and visualizing the Process Ids.

 

When reading an Xdmf file however, it seems that only one processor is loading the data: am I correct in assuming that this is the normal behavior for unstructured datasets ? I used the D3 filter to do some load-balancing after reading the data, but I don't know if this is necessary for subsequent filters to work in parallel ?

 

Also, it seems from the memory inspector that ParaView duplicates the memory footprint of the data on each processor, which quickly becomes a problem for large datasets. Is this the expected behavior, or does it depend on the reader/data format ?

 

Any information on this matter would be greatly appreciated.

Jean-Michel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20160311/c2ab9118/attachment.html>