Too many output of the cube files by mpi |
- Date: 2023/04/13 17:12
- Name: Lingzhi Zhang
<zy38431037@gmail.com>
- Dear OpenMX developer and users,
I'm using openmx to generate STM images.
As I expected, it should give a cube file named as *.pden.cube.
But it actually gives many cube files named *.pden.cube_a. a is a number.
And I find that it meets the number of mpi threads I used.
If I set mpi threads as 100, it gives 100 cube files.
But something strange is that it didn't always give so many cube files.
In some cases, it gives only one cube file although the difference between the cases is just the energy window I set.
I just wonder how should I solve it. I have tried add_gcube, but it failed
with the error infos
[k028230@ohtaka1 -0.5]$ add_gcube ang_xy.pden.cube_000 xy.pden.cube_001 ang_xy.pden.cube
Found a difference in the number of grid on a-axis
Found a difference in the number of grid on b-axis
Found a difference in the number of grid on c-axis
Found a difference in the number of atoms
Found a difference in x-coordinate of the origin
Found a difference in y-coordinate of the origin
Found a difference in z-coordinate of the origin
Found a difference in the vector of a-axis
Found a difference in the vector of b-axis
Found a difference in the vector of c-axis
Segmentation fault (core dumped)
My calculation is conducted in ISSP ohtaka.
My dat file is as below,
stm.dat
<Hubbard.U.values # eV
Ru 1s 0.0 2s 0.0 3s 0.0 1p 0.0 2p 0.0 1d 1.5 2d 0.0 1f 0.0
Cl 1s 0.0 2s 0.0 1p 0.0 2p 0.0 1d 0.0 1f 0.0
C 1s 0.0 2s 0.0 1p 0.0 2p 0.0 1d 0.0
Hubbard.U.values>
#
# SCF or Electronic System
#
scf.XcType GGA-PBE # LDA|LSDA-CA|LSDA-PW|GGA-PBE
scf.SpinPolarization NC # On|Off|NC
scf.SpinOrbit.Coupling On # On|Off, default=off
scf.ElectronicTemperature 0 # default=300 (K)
scf.energycutoff 1200 # default=150 (Ry)
scf.maxIter 2400 # default=40
scf.EigenvalueSolver Band # DC|GDC|Cluster|Band #Check
scf.Kgrid 6 6 1 # means n1 x n2 x n3
scf.Generation.Kpoint regular # regular|MP
scf.Mixing.Type rmm-diish # Simple|Rmm-Diis|Gr-Pulay|Kerker|Rmm-Diisk
scf.Init.Mixing.Weight 0.01 # default=0.30
scf.Min.Mixing.Weight 0.0000000001# default=0.001
scf.Max.Mixing.Weight 0.400 # default=0.40
scf.Mixing.History 50 # default=5
scf.Mixing.StartPulay 50 # default=6
scf.Mixing.EveryPulay 1 # default=6
scf.criterion 1.0e-8 # default=1.0e-6 (Hartree)
scf.lapack.dste dstevx # dstevx|dstedc|dstegr,default=dstevx
#
# Band dispersion
#
Band.dispersion on # on|off, default=off
Band.Nkpath 3
<Band.kpath
40 0.0000000000 0.0000000000 0.0000000000 0.5000000000 0.0000000000 0.0000000000 G M
40 0.5000000000 0.0000000000 0.0000000000 0.3333333333 0.3333333333 0.0000000000 M K
40 0.3333333333 0.3333333333 0.0000000000 0.0000000000 0.0000000000 0.0000000000 K G
Band.kpath>
#
# DOS and PDOS
#
Dos.fileout on # on|off, default=off
Dos.Erange -10.0 10.0 # default = -20 20
Dos.Kgrid 20 20 1 # default = Kgrid1 Kgrid2 Kgrid3
FermiSurfer.fileout on # default = off, on/off
#
# STM
#
partial.charge on # on|off, default=off
partial.charge.energy.window 2.0 # in eV
#
# DFT+U
#
scf.Hubbard.U on # on|off, default=off
scf.DFTU.Type 1 # 1:Simplified(Dudarev)|2:General, default=1
scf.Hubbard.Occupation dual
#
# vdW
#
scf.dftD on
version.dftD 2
DFTD3.damp zero
DFTD.IntDirection 1 1 1
#
# Constraint DFT for NC Spin
#
scf.Constraint.NC.Spin on # on|on2|off, default=off
scf.Constraint.NC.Spin.v 4.0 # default=0.0(eV)
#
# SCF restart
#
scf.restart on
My script is as below
#!/bin/sh
#SBATCH -p F4cpu
#SBATCH -N 4
#SBATCH -n 128
#SBATCH -c 4
#SBATCH -J STM-zz-mono
#SBATCH --mail-type=BEGIN
#SBATCH --mail-type=END
#SBATCH --mail-user=lingzhi@g.ecc.u-tokyo.ac.jp
set -e
source /home/issp/materiapps/intel/openmx/openmxvars-3.9.9-1.sh
energy_list=( -1.5 -1.0 -0.9 -0.5 0.5 1.0 1.5 2.0 2.5 2.8 3.0)
for energy in ${energy_list[@]}
do
awk -v awk_energy=$energy 'FNR==175 {$2=awk_energy} {print $0}' stm.dat > stm.$energy.dat
srun openmx stm.$energy.dat -nt $OMP_NUM_THREADS > stm.$energy.std
echo $energy
prefix=rucl3-GR-ang_xy
mkdir pcharge/$energy
mv $prefix.pden.cube* pcharge/$energy
done
With best regards
Lingzhi Zhang
 | |
This thread is locked.Only browsing is available.
Re: Too many output of the cube files by mpi