Using Paraview in parallel


I am trying to use paraview in a HPC environment. I want to utilize GPUs for rendering and also use multiple cores for performing quicker calculations . I have previously tried running using mpirun -np 24 pvserver, but it didn’t help in calculation speedup. Though paraview is showing the multiple processors (view → memory inspector), it seems that it is using only 1 processor for performing the whole calculation and it is not parallelizing it. When comparing the time taken to perform a calculation in my local laptop to the HPC server, my local paraview is faster. I am not sure if I am loading paraview correctly on my HPC environment. If you can help me troubleshoot this issue, it would be great !


There can be multiple reasons but the most important thing is to diagnose and understand what is taking a long time.

See this blog: