Workflow for large result files

kuhl

Hi all,
we sometimes have the issue that files with solved models (results) are pretty large (up to 10 GB). That makes it difficult to post process results on the laptop (16 GB RAM), because memory is not sufficient to open the file. The work station we use to solve the models is remote and therefore not really suitable for post processicing.
Is there any way to produce a reduced file with results only to be used for post processing?
Regards,
Kuhl

Sergio

You could reduce the number of saved intermediate increments if is a no lineal simulation (maybe leaving only the last one), and also reduce the number of field variables to the one that you really need to postprocess. A last resource would be transform your result to Paraview and postprocess there.

kuhl

Thanks for the advice.
It is linear simulation, so we Will try to reduce the output variables.
I have had no luck with paraview.
Can you help me with the correct workflow for opeding pvd / vtu files?
I get this error message in paraview:
"ERROR: In C:\bbd\2d618e80\build\superbuild\paraview\src\VTK\IO\XML\vtkXMLDataReader.cxx, line 443
vtkXMLUnstructuredGridReader (000001EF62A83260): Cannot read point data array "displacement" from PointData in piece 0. The data array in the element may be too short."

Sergio

I don´t have much experience with Mecway´s internal converter, I use Calmed8 instead, that convert to only one .rmed result file that can be opened with Paraview. I have converted several no lineal results with about 50 increments (up to 20 GB) to one .rmed file and Paraview can manage it razonably well, even create animations.

kuhl

thank you, I'll give that a try