Parallel Meep - any example code that shows linear speedup?

[rcoridan]

I am trying to implement parallel versions of my serial meep code. I am not seeing significant speedup when I am using the parallel pymeep implementation with mpirun on a macbook. I am afraid that I use too much DFT field routines (o get a transmission/reflectance spectra that will affect speedup. Is there some example code that will have significant (even linear) speedup when I add processors?

Thanks for any guidance.