problem with parallel run in ABINIT 6.6.1 version
Posted: Thu Apr 28, 2011 2:32 pm
Dear users
I have a problem on GW correction for Poly para phenylene.
The program stops at the last stage (calculation of GW correction-optdriver: 4) but there is no ERROR message in the log file.
The end of an output file is as below:
" rdkss : COMMENT -
Note that the test on the orthogonalization is not complete
since bands are spread among different processors"
and
My input file is as below:
# Dataset1: usual self-consistent ground-state calculation
prtden1 1 # Print out density
prtpot1 1 # Print out the Hartree potential
prtvxc1 1 # Print out the Hartree potential
nband1 20
nbandkss1 20 # Number of bands in KSS file (the maximum possible)
kssform1 3 # conjugate gradient algorithm
# Calculation of the dielectric matrix
optdriver2 3
getkss2 1
ecuteps2 3.0
ecutwfn2 25.0
awtr2 1
nband2 20
gw_EET2 2 # switch on energy effective technique (EET) by Berger et al. PRB 82, 041103 (2010).
# Calculation of the GW corrections
optdriver3 4
getkss3 1
getscr3 2
nband3 20
ecutwfn3 25.0
ecutsigx3 8.0
gw_EET3 2
nkptgw3 1
kptgw3 0.000 0.000 0.000
bdgw3 1 20
# GW calculation general parameters
ppmodel 2
icutcoul 1
vcutgeo 0 0 1
rcut 15
inclvkb 2
mffmem 1
gwmem 0
gwpara 2
# Definition of the SCF procedure
ecut 25.0
nstep 100
toldfe 1e-10
diemac 2.0
iscf 7
# BZ sampling
kptopt 1
nshiftk 1
shiftk 0.0 0.0 0.0
istwfk *1
ngkpt 1 1 8
symmorphi 0
# Definition of the structure
####################################################################################
acell 35.000000000 35.000000000 8.077578004
rprim 1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
# Definition of the atom types
ntypat 2
znucl 6 1
natom 10
typat 2 1 1 2 1 1 2 1 1 2
xangst
7.655179 0.000000 2.620940888
8.614955 0.000000 2.107050888
9.819599 0.000000 2.829066888
7.655179 0.000000 0.213897888
8.614955 0.000000 0.727791888
11.024243 0.000000 2.107050888
11.984019 0.000000 2.620940888
9.819599 0.000000 0.005775888
11.024243 0.000000 0.727791888
11.984019 0.000000 0.213897888
I don't know what is wrong.
Does anyone know what happened?
Any suggestions will be appreciated.
Peiman
I have a problem on GW correction for Poly para phenylene.
The program stops at the last stage (calculation of GW correction-optdriver: 4) but there is no ERROR message in the log file.
The end of an output file is as below:
" rdkss : COMMENT -
Note that the test on the orthogonalization is not complete
since bands are spread among different processors"
and
My input file is as below:
# Dataset1: usual self-consistent ground-state calculation
prtden1 1 # Print out density
prtpot1 1 # Print out the Hartree potential
prtvxc1 1 # Print out the Hartree potential
nband1 20
nbandkss1 20 # Number of bands in KSS file (the maximum possible)
kssform1 3 # conjugate gradient algorithm
# Calculation of the dielectric matrix
optdriver2 3
getkss2 1
ecuteps2 3.0
ecutwfn2 25.0
awtr2 1
nband2 20
gw_EET2 2 # switch on energy effective technique (EET) by Berger et al. PRB 82, 041103 (2010).
# Calculation of the GW corrections
optdriver3 4
getkss3 1
getscr3 2
nband3 20
ecutwfn3 25.0
ecutsigx3 8.0
gw_EET3 2
nkptgw3 1
kptgw3 0.000 0.000 0.000
bdgw3 1 20
# GW calculation general parameters
ppmodel 2
icutcoul 1
vcutgeo 0 0 1
rcut 15
inclvkb 2
mffmem 1
gwmem 0
gwpara 2
# Definition of the SCF procedure
ecut 25.0
nstep 100
toldfe 1e-10
diemac 2.0
iscf 7
# BZ sampling
kptopt 1
nshiftk 1
shiftk 0.0 0.0 0.0
istwfk *1
ngkpt 1 1 8
symmorphi 0
# Definition of the structure
####################################################################################
acell 35.000000000 35.000000000 8.077578004
rprim 1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0
# Definition of the atom types
ntypat 2
znucl 6 1
natom 10
typat 2 1 1 2 1 1 2 1 1 2
xangst
7.655179 0.000000 2.620940888
8.614955 0.000000 2.107050888
9.819599 0.000000 2.829066888
7.655179 0.000000 0.213897888
8.614955 0.000000 0.727791888
11.024243 0.000000 2.107050888
11.984019 0.000000 2.620940888
9.819599 0.000000 0.005775888
11.024243 0.000000 0.727791888
11.984019 0.000000 0.213897888
I don't know what is wrong.
Does anyone know what happened?
Any suggestions will be appreciated.
Peiman