I recently adapted the pivlab_comandline script in order to run it on our lab cluster with the mcc toolbox.
I have added a script to generate schedduler commands (SLURM for the moment).
The performance are quite interesting, proccessing of 10000 1280x800 images to get 9999 velocity fields takes 5 to 10 minutes on 100 cores.
Are you intersted in adding this facilities in a future version of pivlab?
In this case i can send you the additionnal scripts that i have written
for academic purposes, most of the labs have access to parallel computers which are generally used for numerical simulation.
My personal thinking is that PIV can be used for turbulence research as far as you can converge high order statistics (not only average and rms value).
On my personnal experience during my phd i waste a lot of time processing PIV images on a personal computer. Therefore my opinion is that it could be useful for people who study turbulent and coupled phenomena that require a very large amount of images to process.
By the way PIVlab is fabulously efficient and thank you to make it public. You are of course the one who code so you decide.
Is it possible to use multiple cores on a same computer in parallel using PIVLab? I have 12 cores on my computer (but only 1 GPU) and would love to speed up processing in PIVLab, but don't know where or how to put in such scripts. Any help you can provide would be very helpful, thank you!