ABINIT Discussion Forums

hello
I have an important question about memory
I ran lesson 41 for LaCoO3 with this inputs

# Crystalline LaCoO3(rhombohedral) : optimization of the lattice parameter
# at fixed number of k points and broadening.

#Definition of occupation numbers
occopt 4
tsmear 0.05

#Definition of the unit cell
acell 3*10.163 # This is equivalent to 7.60 7.60 7.60
#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 2
ionmov 3
ntime 10
dilatmx 2.50
ecutsm 0.1

#Definition of the atom types
ntypat 3 # There is only one type of atom
znucl 57 27 8 # 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 10 # There are two atoms
typat 1 1 2 2 3 3 3 3 3 3 # They both are of type 1, that is, Hydrogen
xcart # This keyword indicate that the location of the atoms
0.2500 0.2500 0.2500 # will follow, one triplet of number for each atom
0.7500 0.7500 0.7500 # Triplet giving the cartesian coordinates of atom 1, in Bohr
0.0000 0.0000 0.0000 # Triplet giving the cartesian coordinates of atom 2, in Bohr
0.5000 0.5000 0.5000
0.1982 0.3018 0.7500
0.8018 0.6982 0.2500
0.3018 0.7500 0.1982
0.6982 0.2500 0.8018
0.7500 0.1982 0.3018
0.2500 0.8018 0.6982

angdeg 3*60.798
#Definition of the planewave basis set
ecut 45.0 # Maximal kinetic energy cut-off, in Hartree

#Exchange-correlation functional
ixc 1 # LDA Teter Pade parametrization

#Definition of the k-point grid
kptopt 1 # Option for the automatic generation of k points, taking
# into account the symmetry
ngkpt 2 2 2 #This is a 2x2x2 FCC grid, based on the primitive vectors
nshiftk 1 # of the reciprocal space. For a FCC real space lattice,
# like the present one, it actually corresponds to the
# so-called 4x4x4 Monkhorst-Pack grid, if the following shifts
# are used :
shiftk 0.5 0.5 0.5
# 0.5 0.0 0.0
# 0.0 0.5 0.0
# 0.0 0.0 0.5

#Definition of the SCF procedure
nstep 65 # 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)


and the error said

maxfor

chkdilatmx: ERROR -
The new primitive vectors rprimd (an evolving quantity)
are too large with respect to the old rprimd and the accompanying dilatmx :
this large change of unit cell parameters is not allowed by the present value of dilatmx.
You need at least dilatmx= 3.991223E+01
Action : increase the input variable dilatmx.

leave_new : decision taken to exit ...
1

and when I increase the dilatmx the error said

Biggest array : cg(disk), with 1489.5058 MBytes.
memana : allocated an array of 1489.506 Mbytes, for testing purposes.

memana : ERROR -
Test failed to allocate 4678.064 Mbytes
It is not worth to continue
Action : modify input variable to fit the available memory.
or increase limit on available memory.

leave_new : decision taken to exit ...
1
I don't know what must I do!
please help me sooner
thanks

There are only five no. of atoms in La Cu O3 so why are u going to use no. of atoms as 10?

hello
in rhombohedral phsae LaCoO3 has 10 atoms in unit cell
thanks

1) dilatmx should not be more than 1.1 or so. This is what explodes your memory.

2) if the structure is expanding too much and needs such a large dilatmx, the equilibrium is nowhere near your initial guess (experiment?). Check your pseudopotentials on simpler compounds and make sure you have the correct structure (bohr vs angstroem, cell shape etc...)

3) If other things, such as ecut or nkpt, are not converged, you may get unphysical equilibria and the cell will not converge to anything near your initial guess. Check convergence on all these too.



Matthieu

hello
thank you for your help
I use .pspnc pseudopotentials for my program, can you help me and say if it's right or no?
thanks alot