The ABINIT Users Mailing List ( CLOSED )

[Marcel Mohr <marcel@physik.tu-berlin.de>]

Hello Jim

> I'm running a dataset to calculate phonon response and eigendisplacements.  My
> initial run took 3 hours (on 10 processors), but was a run for only 1 atom, 
> and
> in 1 direction.
>
> I have started another run, again on 10 processors, but this time for all 43
> atoms in all 3 directions.  I obviously understand that this may take
> significantly longer.  so far, it's over 167 hours in.

So I would estimate 3 hours x 43 x 3 = 387 hours.
Be happy that it scales linearly ;)

Cheers Marcel