I have data generated from Converge CFD solver which is about 12 gb in size. I am using Paraview 5.10.1 on mac os monterey 12.5.1. I am facing the following difficulty on both windwos and mac os machines.
- I could see MPI on both the machines but I am unable to see how many cores paraview is using in both the machines,
- I read the data file with convergecfd reader and it takes a few minutes to read the data, when I apply celltopointdata filter it takes a few more minutes to get render the visualization, On top of this if I apply clip filter using either a plane or a box with specified dimensions paraview stops responding and I have to force quit the paraview.
I am not sure what’s happening with the clip filter. Any help is appreciated.
I could see MPI on both the machines but I am unable to see how many cores paraview is using in both the machines,
How do you run ParaView ? ParaView will not start using MPI unless you configure it to do so.
I am running by executing the app icon (clicking the app Icon ) when I see the about paraview in the macos its says MPI enabled.
You are not using MPI then, but anyway, since you are working on a single computer, I dont think you need to use tbh.
Regarding your clip, well clipping can be an extensive operation, did you try letting it run a few hours just to know if it actually finishes ?
I left it the whole night and still, it did not finish the process… its about 8 to 9 hrs duration. By the way, could you help me run the paraview on a multicore processor as well? How do I run praview in parallel on a multicore processor. I also tried running the binary version of paraview on our hpc and it was giving the following error
qt.qpa.plugin: Could not load the Qt platform plugin “xcb” in “” even though it was found.
This application failed to start because no Qt platform plugin could be initialized. Reinstalling the application may fix this problem.
Available platform plugins are: xcb.
error: exception occurred: Subprocess aborted
I have used anaconda-based enviroment, used spack to install paraview on my HPC but none of them worked.
I left it the whole night and still, it did not finish the process… its about 8 to 9 hrs duration.
Then you will need to try and reproduce that issue with a smaller dataset.
ParaView is a multithreaded application and most filter will use all your cores, you can monitor your CPU activity during filter execution to check.
Anyway, here is the doc for running ParaView with MPI:
When using the binary, use
./bin/mpiexec instead of mpirun.
I also tried running the binary version of paraview on our hpc and it was giving the following error
I suppose you HPC has no Xorg server running ? In that case, you want to use our osmesa or egl release: https://www.paraview.org/download/
But paraview version 5.4 is working on the HPC. Since I cannot read .h5 converge cfd files directly on 5.4 version I am unable to use it.