[Paraview] atom/particle data in Xdmf

[Eric E. Monson]

Hey Dan,

I look at data which are time series simulating cells swimming around  
in a 3D volume, so maybe it's similar enough (although I think your  
data sets are much larger than mine).

The grid should have a TopologyType="Polyvertex", and then for me, my  
data sets are saved with the positions of the cells in an nCells x 3  
array for x,y,z, so I use GeometryType="XYZ". My Xdmf files are a bit  
strange because I use hyperslabs to split up an nCells x 4 array by  
column into four separate scalar arrays.

Here's an example with five time steps, which are all stored in the  
same HDF5 file (not attached). You can see that after the first few  
time steps the number of particles changes with each time step in this  
case.

Maybe someone else can advise you the best way to get all your ASCII  
data into HDF5. I would probably use a Python script and the PyTables  
package (which makes accessing the HDF5 easy) to convert the data  
since I've had good luck using it to go the other way (from HDF5 to  
other formats) when needed. I also use this combination to generate  
my .xmf files since my collaborators just dump out HDF5 directly and I  
need to build the XML file afterwards to get it all into ParaView.

Hope this helps,
-Eric

------------------------------------------------------
Eric E Monson
Duke Visualization Technology Group

-------------- next part --------------
A non-text attachment was scrubbed...
Name: TemporalSeries_cells.xmf
Type: application/octet-stream
Size: 12734 bytes
Desc: not available
URL: <http://www.paraview.org/pipermail/paraview/attachments/20080618/a5cfd9b8/attachment.obj>
-------------- next part --------------

On Jun 18, 2008, at 5:16 AM, Dan Lussier wrote:

> Hello,
>
> I am looking to put a large amount of existing data into Xdmf format  
> but I'm new to setting up Xdmf and using them in Paraview so I was  
> hoping for some help in the process.  Here is my situation:
>
> - currently have ASCII data with one system snapshot per file  
> (approx 100-200 MB per snapshot)
> - data set is from atomistic simulation so there is no connectivity  
> between any of the data points
>
> Are there any specific recommendations for grid type, topology,  
> geometry type etc... that would suit my situation?
>
> If anyone has a sample XML file that fits this type of situation and  
> would be willing to share that would be great too!
>
> Many thanks.
>
> Dan
> _______________________________________________
> ParaView mailing list
> ParaView at paraview.org
> http://www.paraview.org/mailman/listinfo/paraview