| I've tried using only kpt parallelism but that returns the same error. Indeed prtdos 3 does not seem to work with any kind of parallel instructions. This is however limited only to the use of PAW potentials - as soon as I use a non PAW potential, the projected density of states is calculated.
Any ideas on this?
Thanks, Sam
On 10 Mar 2009, at 14:45, matthieu verstraete wrote:
Yes - to my knowledge prtopt 3 (and I thought 2 as well) only work with kpt parallelism. Perhaps this has been generalized, but it looks like something is not being initialized. It would need much more involved debugging, but it may get done by someone some day. You can look through the calls in outscfcv.F90 and sub-subroutines if you want to consider generalizing it yourself (shouldn't be too hard, as the bands are treated quite independently).
Matthieu
On Tue, Mar 10, 2009 at 3:28 PM, Sam M. Vinko <Sam.Vinko@physics.ox.ac.uk> wrote: Dear Abinitors,
I've come across a strange problem; I'm trying to compute the projected density of states for Al with the option prtdos 3.
The input file works (i.e. does not return an error and outputs a DOS file) with abinis and abinip for all prtdos values (1, 2, 3) if only one cpu is used. If I use abinip with more than 1 cpu then prtdos options 1 and 2 sill work (and are faster to compute) but option 3 gives an error that crashes the simulation:
strhar : ehart,ucvol= 2.48956015486507132E-003 111.92428125000002
partial_dos_fractions : spaceComm = 0
partial_dos_fractions : rmax = 1.34124102233253313E+219
SPLINE_CUBIC_SET - Fatal error!
The number of knots must be at least 2.
The input value of N = 0
Is there any reason why band parallelization should not work for DOS calculations? Am I missing some important variable here?
I'm using Abinit 5.6.4, gcc/gfortran/openmpi.
Cheers,
Sam
The input file:
---------------START--------------------------------------------------------------------
prtdos 3
#Definition of parallelizations
paral_kgb 1
npband 2
npfft 1
npkpt 1
#Definition of occupation numbers
occopt 7
tsmear 0.05
#Definition of the unit cell
acell 3*7.65 # This is equivalent to 7.65 7.65 7.65
rprim 0.0 0.5 0.5 # FCC primitive vectors (to be scaled by acell)
0.5 0.0 0.5
0.5 0.5 0.0
#Optimization of the lattice parameters
optcell 0
#Other optimizations
wfoptalg 4
fftalg 401
fft_opt_lob 2
nloalg 4
iscf 17
#Definition of the atom types
ntypat 1 # There is only one type of atom
znucl 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. Here, the only type is Aluminum
#Definition of the atoms
natom 1 # There is only one atom per cell
typat 1 # This atom is of type 1, that is, Aluminum
xred # This keyword indicate that the location of the atoms
# will follow, one triplet of number for each atom
0.0 0.0 0.0 # Triplet giving the REDUCED coordinate of atom 1.
#Definition of the planewave basis set
ecut 18.0 # Maximal kinetic energy cut-off, in Hartree
pawecutdg 50.0
nband 60
#Definition of the k-point grid
kptopt 1 # Option for the automatic generation of k points, taking
# into account the symmetry
ngkpt 4 4 4
#Definition of the SCF procedure
nstep 100 # Maximal number of SCF cycles
toldfe 1.0d-6 # Will stop when, twice in a row, the difference
# between two consecutive evaluations of total energy
# differ by less than toldfe (in Hartree)
-------------------END------------------------------------------------------------------------
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Matthieu Verstraete
European Theoretical Spectroscopy Facility (ETSF)
Dpto. Fisica de Materiales,
U. del Pais Vasco,
Centro Joxe Mari Korta, Av. de Tolosa, 72, Phone: +34-943018393
E-20018 Donostia-San Sebastian, Spain Fax : +34-943018390
Mail : matthieu.jean.verstraete@gmail.com
http://www-users.york.ac.uk/~mjv500
|
