For a system of this size, PW calculations are generally slow in the 2017 version, but we are working on improvements for the next major release.
The current calculation does not seem likely to converge, and generally your script looks fine. My first suggestion would be to increase the vacuum a little and center the configuration.
My second suggestion would be to raise the cutoff energy. Our suggested mesh cutoff is a 140 Hartree in LCAO calculations with these pseudopotentials, and you should probably aim for something similar.
I would also like to draw your attention to our small tutorial on the PW calculator, if you are not already familiar with it:
https://docs.quantumwise.com/tutorials/pw_intro/pw_intro.html