SNL: Spyplot reader problem on node 0

This issue was created automatically from an original Mantis Issue. Further discussion may take place here.

When running paraview on very large AMR CTH data (640 million cells) there is a problem with node 0. We're running 128 node server, and all the nodes are using about 200 meg of memory (including node 0) then at the very end of the read node 0 suddenly goes to like 2 gig of memory and then crashes. I will try to get some large data that replicates this problem... in the meantime someone might look to see what node 0 is 'collecting' at the very end that might be memory intensive when working with 400k datasets (AMR grids) and 460M cells.

To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information