ABINIT Discussion Forums

Dear All,
I was trying to do RF (phonon calculations) for Fe32Si31Al ( 64 atoms). I was trying also to utilize the paralleization levels implemented in Abinit.
The job I submitted which has 15 datasets based on 4 4 4 mesh.
I was using 240 processors, these are specified as npband 120 ( nband 240) and npkpt 2 ( nkpt 14 for DATASET1 and 64 for DATASETS2-15).
however, the job ran for three hours and then stopped with this message:

=================================================
[color=#FF0000]getcut: wavevector= 0.0000 0.0000 0.0000 ngfft= 90 90 90
ecut(hartree)= 30.000 => boxcut(ratio)= 2.15148
_pmiu_daemon(SIGCHLD): [NID 03951] [c13-2c2s7n1] [Sat Oct 27 19:27:00 2012] PE RANK 39 exit signal Segmentation fault
_pmiu_daemon(SIGCHLD): [NID 03949] [c13-2c2s6n1] [Sat Oct 27 19:27:00 2012] PE RANK 179 exit signal Segmentation fault
[NID 03951] 2012-10-27 19:27:00 Apid 12443833: initiated application termination
_pmiu_daemon(SIGCHLD): [NID 04017] [c13-2c2s7n3] [Sat Oct 27 19:27:00 2012] PE RANK 82 exit signal Segmentation fault
_pmiu_daemon(SIGCHLD): [NID 04019] [c13-2c2s6n3] [Sat Oct 27 19:27:00 2012] PE RANK 127 exit signal Segmentation fault
_pmiu_daemon(SIGCHLD): [NID 04498] [c11-2c2s6n0] [Sat Oct 27 19:27:00 2012] PE RANK 203 exit signal Segmentation fault
Application 12443833 exit codes: 139
Application 12443833 exit signals: Killed
Application 12443833 resources: utime ~41871s, stime ~103s Rss ~522416 inblocks ~312514 outblocks ~745976[/color]
====================================================

Below is the input file I used
#Fe32Si31A.in
# computation of the phonon band structure and related quantities
#--------------------------------------------------------
ndtset 15
#--------------------------------------------------------
#Set 1 : ground state self-consistency

getwfk1 0 # Cancel default
kptopt1 1 # Option for the automatic generation of k points,
# taking into account the symmetry
# Defintion of the different grids

nqpt1 0 # Cancel default
tolvrs1 1.0d-16 # SCF stopping criterion (modify default)
rfphon1 0 # Cancel default
#-----------------------------------------------------------------
#Q vectors for all datasets

#Complete set of symmetry-inequivalent qpt chosen to be commensurate
# with kpt mesh so that only one set of GS wave functions is needed.
#Generated automatically by running GS calculation with kptopt=1,
# nshift=0, shiftk=0 0 0 (to include gamma) and taking output kpt set
# file as qpt set. Set nstep=1 so only one iteration runs.

nqpt 1 # One qpt for each dataset (only 0 or 1 allowed)
# This is the default for all datasets and must
# be explicitly turned off for dataset 1.

qpt2 0.00000000E+00 0.00000000E+00 0.00000000E+00
qpt3 2.50000000E-01 0.00000000E+00 0.00000000E+00
qpt4 5.00000000E-01 0.00000000E+00 0.00000000E+00
qpt5 2.50000000E-01 2.50000000E-01 0.00000000E+00
qpt6 5.00000000E-01 2.50000000E-01 0.00000000E+00
qpt7 -2.50000000E-01 2.50000000E-01 0.00000000E+00
qpt8 2.50000000E-01 5.00000000E-01 0.00000000E+00
qpt9 5.00000000E-01 5.00000000E-01 0.00000000E+00
qpt10 2.50000000E-01 2.50000000E-01 2.50000000E-01
qpt11 5.00000000E-01 2.50000000E-01 2.50000000E-01
qpt12 -2.50000000E-01 2.50000000E-01 2.50000000E-01
qpt13 5.00000000E-01 5.00000000E-01 2.50000000E-01
qpt14 -2.50000000E-01 5.00000000E-01 2.50000000E-01
qpt15 5.00000000E-01 5.00000000E-01 5.00000000E-01

#--------------------------------------------------------
# note that for metals we do not have to take into account the coupling
# between atomic displacments and the homogenous electric feild (exists
# in case of polar insulators). that is, we do nothave to do d/dk perturbation.
#that is data set 2 and 3 are not required.
#Also do not forget to consider occopt and tsmear
#
##Set 2 : Response function calculation of d/dk wave function

# iscf2 -3 # Need this non-self-consistent option for d/dk
# kptopt2 2 # Modify default to use time-reversal symmetry
# rfphon2 0 # Cancel default
# rfelfd2 2 # Calculate d/dk wave function only
# tolwfr2 1.0d-20 # Use wave function residual criterion instead

#-------------------------------------------------------------------
##Set 3 : Response function calculation of Q=0 phonons and electric field pert.

# getddk3 2 # d/dk wave functions from last dataset
# kptopt3 2 # Modify default to use time-reversal symmetry
# rfelfd3 3 # Electric-field perturbation response only

#-----------------------------------------------------------------------
#Sets 2-15 :Finite-wave-vector phonon calculations (defaults for all datasets)

getwfk 1 # Use GS wave functions from dataset1
kptopt 3 # Need full k-point set for finite-Q response
rfphon 1 # Do phonon response
rfatpol 1 64 # Treat displacements of all atoms
rfdir 1 1 1 # Do all directions (symmetry will be used)
tolvrs 1.0d-8 # This default is active for sets 3-25
#---------------------------------------------------------------------------

#Definition of the unit cell

#ndtset 11
acell 3*16.966 # experimental lattice paramtere is used
rprim 1.0 0.0 0.0 # In lessons 1 and 2, these primitive vectors
0.0 1.0 0.0 # (to be scaled by acell) were 1 0 0 0 1 0 0 0 1
0.0 0.0 1.0 # that is, the default.
#--------------------------------------------------------
#Definition of the atom types
ntypat 3 # There are three types of atoms
znucl 26 14 13 # The keyword "znucl" refers to the atomic number of the
# possible type(s) of atom. The pseudopotential(s)
# mentioned in the "files" file must correspond
# to the type(s) of atom.
#-------------------------------------------------------
#Definition of the atoms
natom 64 # There are 64 atoms
typat 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 2 2 2 2 2 2 2 2 2 2 2 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

xred # relaxed atomic positions, using 4 4 4 k-mesh
# and 30H ecut, ixc=1 dimac =9.0. Based on relaxed
# atomic positions of 1 unitcell, using exp lattice
# parameter a=8.483Bohr (888 and 30)

0.066796641683 0.066796641683 0.066796641683
....
....
0.830376381928 0.580518851632 0.169809017731
#--------------------------------------------------------
#Definition of the planewave basis set
ecut 30.0 # Maximal kinetic energy cut-off, in Hartree
ixc 1
nband 240
paral_kgb 1
npkpt 2
npband 120
occopt 3 # Fermi-Dirac
tsmear 0.008 # ~0.2 eV, Default 0.04Ha
timopt 2 # for parallel excution
#Definition of the k-point grid

ngkpt 4 4 4
nshiftk 1
shiftk 0.0 0.0 0.0

#getwfk -1 # this is to speed up the calcualtions

#-----------------------------------------------------------
#Definition of the SCF procedure
iscf 5
nstep 100 # Maximal number of SCF cycles
#toldfe 1.0d-6 # no longer needed, toldff is used above
# Will stop when, twice in a row, the difference
# between two consecutive evaluations of total energy
# differ by less than toldfe (in Hartree)
diemac 9.0 # Although this is not mandatory, it is worth to
# precondition the SCF cycle. The model dielectric
# function used as the standard preconditioner
# is described in the "dielng" input variable section.



I look forward to have your comments and suggestions.

Thanks
IYAD

I am pretty sure that the kgb parallelism has not been implemented yet in response function calculations, but I may be wrong on this.