Dear All
I am working on a FC calculation on a large system. I know that we can
split the calculation over every atoms in the unit cell. But in my case
the FC calculation stop at the 6th step because of the walltime, and I
don't want to repeat the previous 5 steps with more CPU cores again
(usesavedata has been used).
Could anyone please tell me if it is possible that we do only one
displacement on a certain direction (let say +z direction) of one atom
to calculate the force constants, and then combine the FC file together
with the previous 5 steps of the same atom. Thanks.
Regards,
Tao
- [SIESTA-L] FC calculation continuation T. Liu
-