[Pw_forum] relax question
zhentang wang
wangzhentang at gmail.com
Wed Dec 19 04:32:13 CET 2007
Dear pwscf users,
I am runing in parallel a relax conculation ,and there are some problems
to ask.
cluster Set up
pwscf versin: 3.2.3
compiler: ifort I_fc_p_10.0.23 mkl9.1.023
mpi: moich2-1.0.06
system : cluster xeon 5345 (64bit)
Input
&control
calculation='relax',
restart_mode='from_scratch',
prefix='GaMnAs',
pseudo_dir = '/home/ztwang/PP',
outdir='tmp',
tprnfor=.true.,
disk_io='high',
verbosity='high',
/
&system
ibrav=1, celldm(1) = 21.0,
nat=65, ntyp=3,
ecutwfc=60,
ecutrho=360.0,
starting_magnetization(1)=0.3,
starting_magnetization(2)=0.0,
nspin=2,
occupations='smearing',
degauss=0.02,
smearing='cold',
/
&electrons
mixing_beta = 0.5
conv_thr = 1.0d-6
mixing_mode='local-TF'
/
&ions
//
ATOMIC_SPECIES
Mn 54.938049 MnPZus_15e.vdb.DB.UPF
Ga 69.723 Ga.pz-bhs.UPF
As 74.92160 As.pz-bhs.UPF
ATOMIC_POSITIONS
Mn 0.250000000 0.250000000 0.250000000
Ga 0.000000000 0.000000000 0.000000000
Ga 0.000000000 0.250000000 0.250000000
Ga 0.250000000 0.000000000 0.250000000
Ga 0.250000000 0.250000000 0.000000000
As 0.125000000 0.125000000 0.125000000
As 0.125000000 0.375000000 0.375000000
As 0.375000000 0.125000000 0.375000000
As 0.375000000 0.375000000 0.125000000
Ga 0.000000000 0.000000000 0.500000000
Ga 0.000000000 0.250000000 0.750000000
Ga 0.250000000 0.000000000 0.750000000
Ga 0.250000000 0.250000000 0.500000000
As 0.125000000 0.125000000 0.625000000
As 0.125000000 0.375000000 0.875000000
As 0.375000000 0.125000000 0.875000000
As 0.375000000 0.375000000 0.625000000
Ga 0.000000000 0.500000000 0.000000000
Ga 0.000000000 0.750000000 0.250000000
Ga 0.250000000 0.500000000 0.250000000
Ga 0.250000000 0.750000000 0.000000000
As 0.125000000 0.625000000 0.125000000
As 0.125000000 0.875000000 0.375000000
As 0.375000000 0.625000000 0.375000000
As 0.375000000 0.875000000 0.125000000
Ga 0.000000000 0.500000000 0.500000000
Ga 0.000000000 0.750000000 0.750000000
Ga 0.250000000 0.500000000 0.750000000
Ga 0.250000000 0.750000000 0.500000000
As 0.125000000 0.625000000 0.625000000
As 0.125000000 0.875000000 0.875000000
As 0.375000000 0.625000000 0.875000000
As 0.375000000 0.875000000 0.625000000
Ga 0.500000000 0.000000000 0.000000000
Ga 0.500000000 0.250000000 0.250000000
Ga 0.750000000 0.000000000 0.250000000
Ga 0.750000000 0.250000000 0.000000000
As 0.625000000 0.125000000 0.125000000
As 0.625000000 0.375000000 0.375000000
As 0.875000000 0.125000000 0.375000000
As 0.875000000 0.375000000 0.125000000
Ga 0.500000000 0.000000000 0.500000000
Ga 0.500000000 0.250000000 0.750000000
Ga 0.750000000 0.000000000 0.750000000
Ga 0.750000000 0.250000000 0.500000000
As 0.625000000 0.125000000 0.625000000
As 0.625000000 0.375000000 0.875000000
As 0.875000000 0.125000000 0.875000000
As 0.875000000 0.375000000 0.625000000
Ga 0.500000000 0.500000000 0.000000000
Ga 0.500000000 0.750000000 0.250000000
Ga 0.750000000 0.500000000 0.250000000
Ga 0.750000000 0.750000000 0.000000000
As 0.625000000 0.625000000 0.125000000
As 0.625000000 0.875000000 0.375000000
As 0.875000000 0.625000000 0.375000000
As 0.875000000 0.875000000 0.125000000
Ga 0.500000000 0.500000000 0.500000000
Ga 0.500000000 0.750000000 0.750000000
Ga 0.750000000 0.500000000 0.750000000
Ga 0.750000000 0.750000000 0.500000000
As 0.625000000 0.625000000 0.625000000
As 0.625000000 0.875000000 0.875000000
As 0.875000000 0.625000000 0.875000000
As 0.875000000 0.875000000 0.625000000
K_POINTS {automatic}
2 2 2 1 1 1
error information was found, it displyed
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
End of self-consistent calculation
convergence NOT achieved, stopping
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
And then I change ecutwfc=40,ecutrho=200.0
error information was found, it displyed
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
%%%%%%%%%%%%%%%%%%%%%%%%%%
from cdiaghg : error # 679
info =/= 0
%%%%%%%%%%%%%%%%%%%%%%%%%%%
stopping ...
rank 0 in job 16 node01_33013 caused collective abort of all ranks
exit status of rank 0: killed by signal 9
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
And then I change ecutwfc=22,ecutrho=100.0,there is also no problem.but I
think total energy may be not accurate.
I wonder if this is a problem with the system or a problem of the
grogram(MPI) or a problem with input setting.
Thank you very much in advance,
Best regards
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20071219/eac81591/attachment.html>
More information about the users
mailing list