PostProcessing Code_Aster FEM results on cloudHPC
You have run your simulation with code_aster FEM on the cloud and it is converged and now you need to post-process the results.
The results files are several hundred GB and you don’t want to saturate your internet bandwith.
This could be a tipical cloud computing scenario.
You absolutely need to understand how post-processing is done within the CloudHPC platform, service provided bye CFD FEA SERVICE.
IMPORTANT: you will first have launch the simulation inside CloudHPC. If you want to know how to do it, look at this link.
Run a static instance for ubuntu as first step (see this link).
Go to Simulation and click on the plus icon.
Now set the vCPU, the core type (highmem 8GB) and select the folder where the results are located.
In the software list, select UBUNTU 2004.
Click save to start the analysis.
Open The Results
Click on the view icon to open the simulation,
Wait for the remote desktop video screen to initialize.
Click on the icon to expand the view.
Now find the program Salome to open Paravis at its inside. Paravis is an upgraded version of Paraview.
Open Paravis and click on the filter column.
Import the results by right clicking and selecting the results file (for code_aster the .med file) and click OK. Obviously this means you have previously run a Code_Aster FEM analysis before actually doing this step.
Change the results and select DEPL as the reference results.
To add the most common filters right click on the top bar and add common used to the item.
A new interface will appear. Now you can set the filter that you want.
We have applied a WARPBYVECTOR filter to see the deformation shape.
When you have finish the postprocessing, it is import to stop the simulation. In the Simulation windows, click on the stop button.
Now set signal as HARD and click Conferm.
Why Post-Processing on CLOUD_HPC
Sometimes the solutions of our simulations are very large (high number of nodes and elements).
In this case it is better to carry out the postprocessing directly on the cloud.