ABINIT Discussion Forums

Hi all,

I'm trying to do a nonlinear response calculation for a magnetic insulating system, namely NiO. For this I followed the tutorial and changed some_path/tutorespfn/Input/tnlo_2.in according to my needs. I use abinit-6.2.1.

But right after starting with the fifth dataset I get some error, whose source I do not understand. I attach the input file. Output and log file are too big to attach them here. Here are the last lines of them:


output:
getshell : finite difference formula of Marzari and Vanderbilt
(see Marzari and Vanderbilt, PRB 56, 12847 (1997), Appendix B)

number of first neighbours : 2
weight : 1411.3440000

number of second neighbours : 6
weight : 282.2688000


log:
...
isppol, ikpt, nband_k= 2 106 16
isppol, ikpt, nband_k= 2 107 16
isppol, ikpt, nband_k= 2 108 16
-P-0000 leave_test : synchronization done...
kpgio: loop on k-points done in parallel
-P-0000
-P-0000 leave_new : decision taken to exit ...


Of course I can give more information. If you need them, just ask.
It would be great if someone could help me.

Thanks.
Silvia Bahmann

Dear Silvia,

I had a similar experience not long ago when I calculated the Raman spectra of an antimagnetic system. After some tests, I realized that this part of code (at least for the 3rd order perturbation of Raman spectra) is not well parallelized. When I specify one core for this part, it works. Luckily this part does not need much time to finish.

Sincerely,
Guangfu Luo

Hello - please be careful - I am not certain either of these is coded spin polarized - Fabiana DaPieve was working on the implementation last year. You should check the release notes and documentation carefully. Xavier is the final source to know if this project is in production.

Matthieu