[QE-users] problem whem more than 1000 k-points

[aritz leonardo liceranzu]

Yes I think so. It was a while ago and I remember that it drove me  
crazy. Pw.x would start normally but it would stop without crashing,  
like waiting for input. I remember that I fixed it as I have  
explained. By the way, it also happened to me  that the problem would  
show only if I exceeded a list with more than 1000 points.



Quoting Paolo Giannozzi <p.giannozzi at gmail.com>:

> Let me understand: are you saying that there might be a problem with the
> so-called "here document" syntax:
>
> pw.x < EOF
> long input here
> EOF
>
> if the long input is too long? funny
>
> Paolo
>
> On Thu, Aug 20, 2020 at 11:53 AM aritz leonardo liceranzu
> <aritz.leonardo at ehu.eus> wrote:
>
>> Dear Fanhao
>>
>> I have encountered the same problem and the solution is easy. It
>> doesn't have anything to do with QE.
>>
>> Simply create an independent nscf input file that contains the big
>> list of kpoints. Do not create the nscf input file within the same job
>> submission script using EOF.
>>
>> Then simply type pw.x < input and it should work.
>>
>> Best
>> Aritz
>>
>>
>>
>> Quoting Fanhao Jia <18329030072 at 163.com>:
>>
>> > Dear developers,
>> > I am doing a nscf calculation of GaAs with so many k-points.
>> > I can run  the calculation if the number of k-points is 1000, and it
>> > is very fast to finish.
>> > But when I increase to 12*12*12=1728 k-points, the  program does not
>> > really run, there is no output or error.
>> > I have increased the nkp in Module/parameter.f90  to 100000,  it
>> > doesn't help. Increasing cpus still doesn't help.
>> > I have attached my input. If you have any idea, please let me know.
>> > Best,
>> > Fanhao
>> >
>> >
>> > &control
>> >    calculation = 'bands'
>> >    pseudo_dir = '../'
>> > /
>> > &system
>> >    ecutwfc = 60.0
>> >    ibrav = 0
>> >    nat = 2
>> >    ntyp = 2
>> >    nbnd=16
>> > /
>> > &electrons
>> >    electron_maxstep = 100
>> >    conv_thr = 1e-10
>> >    diago_full_acc = .true.
>> >    diago_david_ndim = 4
>> >    mixing_ndim = 8
>> >    mixing_mode = 'plain'
>> >    diagonalization = 'david'
>> >    mixing_beta = 0.7
>> > /
>> > CELL_PARAMETERS angstrom
>> >    0.0 2.82675 2.82675
>> >    2.82675 0.0 2.82675
>> >    2.82675 2.82675 0.0
>> > ATOMIC_SPECIES
>> >    Ga 69.723 'Ga.UPF'
>> >    As 74.9216 'As.UPF'
>> > ATOMIC_POSITIONS crystal
>> >    Ga 0.0 0.0 0.0
>> >    As 0.25 0.25 0.25
>> > K_POINTS crystal
>> >   1728
>> >   0.000000000   0.000000000  0.000000000  1.0
>> >   0.000000000   0.000000000  0.083333333  1.0
>> >   0.000000000   0.000000000  0.166666667  1.0
>> >   0.000000000   0.000000000  0.250000000  1.0
>> >   0.000000000   0.000000000  0.333333333  1.0
>> >
>> >
>> >
>> >
>> > | |
>> > 贾帆豪
>> > |
>> > |
>> > 18329030072 at 163.com
>> > |
>> > 签名由网易邮箱大师定制
>>
>>
>>
>>
>> ======================================================
>> Aritz Leonardo Liceranzu
>> Department of Applied Physics II,
>> Faculty of Science and Technology,
>> University of the Basque Country (UPV/EHU)
>> Bº Sarriena s/n, 48940 Leioa, Spain
>>
>> Phone: +34-946015338,           Fax: +34-946013500
>> Mail: aritz.leonardo at ehu.es     https://aritzleonardo.wordpress.com/
>> ======================================================
>>
>> _______________________________________________
>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
>> users mailing list users at lists.quantum-espresso.org
>> https://lists.quantum-espresso.org/mailman/listinfo/users
>
>
>
> --
> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222




======================================================
Aritz Leonardo Liceranzu
Department of Applied Physics II,
Faculty of Science and Technology,
University of the Basque Country (UPV/EHU)
Bº Sarriena s/n, 48940 Leioa, Spain

Phone: +34-946015338,           Fax: +34-946013500
Mail: aritz.leonardo at ehu.es     https://aritzleonardo.wordpress.com/
======================================================