*** Answer copy-pasted from email exchange ***
Hi Asif,
Nanodcal parallelizes with both MPI and threading. There are no MPI runtimes we support on windows and the speed gain on Linux is negligible for single computer calculations (MPI is for multi-computer parallelism).
The number you see is for the MPI parallelism, the code should still be multi-threaded. Let me know if you observe that the CPU usage is limited to one core (over all the simulation since some parts are not parallelisable and single threaded) but that should not be the case.
If you have a cluster you wish to do calculations on, I suggest using device studio to build the experiment. You can then transfer the work folder (e.g. with scp) to the cluster and run it there. You can find the parallel installation instructions at the following link. https://docs.nanoacademic.com/nanodcal/installation/installation/#installation-of-parallel-environment
Note that this will require a MATLAB license.
I will ask that the setting be clarified to better communicate this in the GUI.
Thank you,
Guillaume
