[Paraview] 400GB dataset out of memory on 1k cores?

[Anton Shterenlikht]

>From berk.geveci at kitware.com Tue Sep 22 19:18:53 2015
>
>Hmmm. We are not exactly on the same wavelength :-) When you run a
>threshold filter, you indeed get a volume with all the cells of a given
>grain id. However, when you show it as a Surface (the default mode),
>ParaView extracts the outer surface of that volume. Which ends up being
>equivalent to showing all the interfaces between grains. To actually see
>_all_ of the elements, you need volume rendering. Can you send me a small
>dataset so I can demonstrate the difference in some pictures.

Yes yes, you are right of course!
This is what I meant.
I don't want volume rendering, but
the surfaces of all volumes (clusters, blobs) of cells
of certain value.

I put a small xz-compressed dataset here:
http://eis.bris.ac.uk/~mexas/0.txz

This is a raw binary file
with extents 312, 312, 310.
I'm sorry I haven't had the time to make a really small dataset.
Let me know if this is too big.

I read the data as

ext1    = 312
ext2    = 312
ext3    = 310
ffile   = <whatever you save it as>
from paraview.simple import *
cracks = ImageReader( FilePrefix= ffile )
cracks.DataExtent=[ 0, ext1-1, 0, ext2-1, 0, ext3-1 ]
cracks.DataByteOrder = 'LittleEndian'
cracks.DataScalarType = 'int'

Many thanks for your help

Anton