Skip to Content.
Sympa Menu

forum - Re: [abinit-forum] Need helps on GW calculations

forum@abinit.org

Subject: The ABINIT Users Mailing List ( CLOSED )

List archive

Re: [abinit-forum] Need helps on GW calculations


Chronological Thread 
  • From: "D. R. Hamann" <drhamann@mat-simresearch.com>
  • To: forum@abinit.org
  • Subject: Re: [abinit-forum] Need helps on GW calculations
  • Date: Fri, 02 May 2008 07:37:45 -0400
Dear ธนูสิทธิ์ บุรินทร์ประโคน,

This problem seems to have been around for a while without getting fixed in any official release.  I'm pasting some correspondence I saved from the "developer" mailing list a while ago.  I've had a hack from the author of this running successfully in 5.4.3 for a while.

Don Hamann

Riad:
    I have a version of outkss.f90 that I modified some time ago, which
is compatible with Abinit 5.1.2. It does the following:

1. kssform=1, direct diagonalization with parallelization in k-points
2. kssform=3, write out .kss file based on previous KS calculations with
nband>=nbandkss. It won't crash in parallel version.

    It is well tested in my work. But I have no idea about how to
implement it into the last version. If anybody has an interest to take a
look, I can send the code.

    Best
    Deyu
    
On Tue, 2007-06-26 at 20:15 +0200, Riad Shaltaf wrote:

> Thanks Matteo for your reply however I was thinking that modification of
> the "outkss.f90" to count for the writing of the KSS file under this
> option shall be done. Since the code just crashes at the moment the
> processors achieve the convergence as we all know, I think the
> modification (in principle) shall not be complicated. 
> 
> So I just wanted to raise this issue in case there is any volunteer to
> code it..
> 
> Riad 
> 
> 
> On Tue, 2007-06-26 at 19:25 +0200, Matteo Giantomassi wrote:
  
> > On Tue, 26 Jun 2007, Riad Shaltaf wrote:
> > 
    
> > > Dear Fabien,
> > >
> > > I agree with you however the drawback of "kssform 3" is that it still
> > > not possible to do the calculation in parallel.
> > >
> > > As i can see, people have started using the GW code in heavy stuff like
> > > nanotubes/wires so may be it is the time to parallelize that option..
      
> > 
> > The problem in using kssform==3 in parallel consists in the fact that 
> > the code crashes when different processors try to write wavefunctions 
> > at different k-points on the KSS file.
> > 
> > A possible workaround could be running the GS or NSCF calculation using
> > the parallel version of the code in order to obtain very well converged 
> > wavefunctions for all the bands and k-points you are planning to use in the GW part. 
> > Then you can simply run a sequential job restarting from the converged WFK file 
> > using kssform 3 to output the KSS file.
> > 
> > Concerning the parallel run it's worth to say that the default
> > parallelism is over k-points. If the system under study is 0-dimensional
> > (i.e atoms, molecules, clusters) the parallelization over k-points is
> > obviously useless. In this case one could try to do the GS or NSCF
> > calculation using the parallelism over bands/FFT. 
> > I've never tried, but I don't see any reason why it should not work.
> > 
> > Hope this helps.
> > Best Regards
> > Matteo Giantomassi
> > 
    
> > >
> > > On Tue, 2007-06-26 at 13:38 +0200, Fabien Bruneval wrote:
      
> > >> Dear Shing,
> > >>
> > >> I think you have a problem of memory. You attempt to calculate and then
> > >> diagonalize the Hamiltonian on the plane-wave basis set. In your system,
> > >> you consider 123,000 PW. The <G|H|G'> is therefore a 123,000 x 123,000
> > >> matrix, i.e. 242 Gbytes.
> > >>
> > >> I advise you to study the possibility to use "kssform 3". Furthermore,
> > >> if you intend to perform GW calculation on this nanostructure, I have to
> > >> tell you that the abinit GW code is not able to treat this anisotropic
> > >> case correctly.
> > >>
> > >> Regards,
> > >>
> > >> Fabien
> > >>
> > >>
> > >>
> > >>
> > >> kohws@ihpc.a-star.edu.sg wrote:
        
> > >>> Hi,
> > >>>
> > >>> I am simulating a Silicon nanowire structure with hydrogen termination with abinit 5.2.3 and using the tutorial example as a reference. However, my job always get terminated after the calculation of the KSS band using dataset 1 when it is calculating the <G|H|G'> elements. As shown in the log file (after the input file) below. Please advise.
> > >>>
> > >>> Thanks.
> > >>>
> > >>> Best Regards
> > >>> Wee Shing
> > >>>
> > >>> -------------------------------------------------------------------------------
> > >>>
> > >>> INPUT FILE
> > >>> -----------
> > >>> ndtset      2
> > >>>
> > >>> # Definition of parameters for the calculation of the KSS file
> > >>> nbandkss1   50         # Number of bands in KSS file (-1 means the maximum possible)
> > >>> nband1      120 1       # Number of (occ and empty) bands to be computed
> > >>> istwfk1     4*1
> > >>>
> > >>> # Calculation of the screening (epsilon-1 matrix)
> > >>> optdriver2  3        # Screening calculation
> > >>> getkss2     50       # Obtain KSS file from previous dataset
> > >>> nband2      50 1      # Bands to be used in the screening calculation
> > >>> ecutwfn2    3.1      # Cut-off energy of the planewave set to represent the wavefunctions
> > >>> ecuteps2    4.6      # Cut-off energy of the planewave set to represent the dielectric matrix
> > >>> ppmfrq2    18.7 eV   # Imaginary frequency where to calculate the screening
> > >>>
> > >>> acell 100.0  100.0  7.31
> > >>> xcart    4.1147688868E+01  4.9999997369E+01  1.8316666044E+00    # Si
> > >>>          4.3697018999E+01  4.6422110607E+01  1.8282082001E+00    # Si
> > >>>          4.3697018234E+01  5.3577885430E+01  1.8282081932E+00    # Si
> > >>>          ..................
> > >>>          ..................
> > >>>          ..................
> > >>>
> > >>> ntypat 2
> > >>> znucl 14 1
> > >>>
> > >>> natom 28
> > >>> typat 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
> > >>> nkpt 4
> > >>> shiftk  0.0  0.0 -0.5
> > >>> kpt   0.0  0.0 -0.125
> > >>>       0.0  0.0 -0.375
> > >>>       0.0  0.0  0.125
> > >>>       0.0  0.0  0.375
> > >>> symmorphi 0
> > >>> ecut 10.0
> > >>> nstep   200
> > >>> toldfe  1.0d-6
> > >>> diemac  12.0
> > >>>
> > >>> -------------------------------------------------------------------------------
> > >>> LOG FILE
> > >>> -----------
> > >>>   ABINIT
> > >>>
> > >>>   Give name for formatted input file:
> > >>> Si.in
> > >>>   Give name for formatted output file:
> > >>> Si.out
> > >>>   Give root name for generic input files:
> > >>> Sii
> > >>>   Give root name for generic output files:
> > >>> Sio
> > >>>   Give root name for generic temporary files:
> > >>> Si
> > >>> -P-0000  leave_test : synchronization done...
> > >>> -P-0001  leave_test : synchronization done...
> > >>> -P-0002  leave_test : synchronization done...
> > >>> -P-0003  leave_test : synchronization done...
> > >>>
> > >>> .Version 5.2.3  of ABINIT
> > >>> .(MPI version, prepared for a powerpc_aix5.2.0.0_ibm computer)
> > >>>
> > >>>  Copyright (C) 1998-2006 ABINIT group .
> > >>>  ABINIT comes with ABSOLUTELY NO WARRANTY.
> > >>>  It is free software, and you are welcome to redistribute it
> > >>>  under certain conditions (GNU General Public License,
> > >>>  see ~abinit/COPYING or http://www.gnu.org/copyleft/gpl.txt).
> > >>>
> > >>>  ABINIT is a project of the Universite Catholique de Louvain,
> > >>>  Corning Inc. and other collaborators, see ~abinit/doc/developers/contributors.
> > >>>  Please read ~abinit/doc/users/acknowledgments.html for suggested
> > >>>  acknowledgments of the ABINIT effort.
> > >>>  For more information, see http://www.abinit.org .
> > >>>
> > >>> .Starting date : Tue 26 Jun 2007.
> > >>>
> > >>> - input  file    -> Si.in
> > >>> - output file    -> Si.out
> > >>> - root for input  files -> Sii
> > >>> - root for output files -> Sio
> > >>>
> > >>>  instrng :   108 lines of input have been read
> > >>>
> > >>>  iofn2 : Please give name of formatted atomic psp file
> > >>>  iofn2 : for atom type   1 , psp file is 14si.pspnc
> > >>>   read the values zionpsp=  4.0 , pspcod=   1 , lmax=   2
> > >>>
> > >>>  iofn2 : Please give name of formatted atomic psp file
> > >>>  iofn2 : for atom type   2 , psp file is 1h.pspnc
> > >>>   read the values zionpsp=  1.0 , pspcod=   1 , lmax=   0
> > >>>
> > >>>  iofn2 : deduce mpsang  =   3, n1xccc  =2501.
> > >>> -P-0000  leave_test : synchronization done...
> > >>>
> > >>>  invars1m : enter jdtset=     1
> > >>>  ingeo : takes atomic coordinates from input array xcart
> > >>>
> > >>>  symanal : COMMENT -
> > >>>   The Bravais lattice determined only from the primitive
> > >>>   vectors, bravais(1)=  4, is more symmetric
> > >>>   than the real one, iholohedry=  1, obtained by taking into
> > >>>   account the atomic positions.
> > >>>  symspgr : the symmetry operation no.   1 is the identity
> > >>>  symspgr : spgroup=   1  P1   (=C11)
> > >>>  inkpts: Sum of    4 k point weights is    4.000000
> > >>>    With present occopt=   1 , renormalize it to one
> > >>>
> > >>>  inkpts : istwfk preprocessed, gives following first values (max. 6): 1 1 1 1
> > >>>   distrb2: enter
> > >>>   mpi_enreg%parareel= 0
> > >>>   mpi_enreg%paralbd= 0
> > >>>   mpi_enreg%paral_compil_respfn= 0
> > >>>   distrb2: exit
> > >>>  invars1: mkmem  undefined in the input file. Use default mkmem  = nkpt
> > >>>  invars1: With nkpt_me=    1 and mkmem  =     4, ground state wf handled in core.
> > >>>  Resetting mkmem  to nkpt_me to save memory space.
> > >>>  invars1: mkqmem undefined in the input file. Use default mkqmem = nkpt
> > >>>  invars1: With nkpt_me=    1 and mkqmem =     4, ground state wf handled in core.
> > >>>  Resetting mkqmem to nkpt_me to save memory space.
> > >>>  invars1: mk1mem undefined in the input file. Use default mk1mem = nkpt
> > >>>  invars1: With nkpt_me=    1 and mk1mem =     4, ground state wf handled in core.
> > >>>  Resetting mk1mem to nkpt_me to save memory space.
> > >>>
> > >>>  invars1m : enter jdtset=     2
> > >>>  ingeo : takes atomic coordinates from input array xcart
> > >>>
> > >>>  symanal : COMMENT -
> > >>>   The Bravais lattice determined only from the primitive
> > >>>   vectors, bravais(1)=  4, is more symmetric
> > >>>   than the real one, iholohedry=  1, obtained by taking into
> > >>>   account the atomic positions.
> > >>>  symspgr : the symmetry operation no.   1 is the identity
> > >>>  symspgr : spgroup=   1  P1   (=C11)
> > >>>  inkpts: Sum of    4 k point weights is    4.000000
> > >>>    With present occopt=   1 , renormalize it to one
> > >>>
> > >>>  inkpts : istwfk preprocessed, gives following first values (max. 6): 1 1 1 1
> > >>>   distrb2: enter
> > >>>   mpi_enreg%parareel= 0
> > >>>   mpi_enreg%paralbd= 0
> > >>>   mpi_enreg%paral_compil_respfn= 0
> > >>>   distrb2: exit
> > >>>  invars1: mkmem  undefined in the input file. Use default mkmem  = nkpt
> > >>>  invars1: With nkpt_me=    1 and mkmem  =     4, ground state wf handled in core.
> > >>>  Resetting mkmem  to nkpt_me to save memory space.
> > >>>  invars1: mkqmem undefined in the input file. Use default mkqmem = nkpt
> > >>>  invars1: With nkpt_me=    1 and mkqmem =     4, ground state wf handled in core.
> > >>>  Resetting mkqmem to nkpt_me to save memory space.
> > >>>  invars1: mk1mem undefined in the input file. Use default mk1mem = nkpt
> > >>>  invars1: With nkpt_me=    1 and mk1mem =     4, ground state wf handled in core.
> > >>>  Resetting mk1mem to nkpt_me to save memory space.
> > >>>
> > >>>  DATASET    1 : space group P1 (#  1); Bravais aP (primitive triclinic)
> > >>>  inkpts: Sum of    4 k point weights is    4.000000
> > >>>    With present occopt=   1 , renormalize it to one
> > >>>
> > >>>  inkpts : istwfk preprocessed, gives following first values (max. 6): 1 1 1 1
> > >>>  chkneu : initialized the occupation numbers for occopt=    1
> > >>>     spin-unpolarized case :
> > >>>   2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00
> > >>>   2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00
> > >>>   2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00
> > >>>   2.00  2.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00
> > >>>   0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00
> > >>>   0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00
> > >>>   0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00
> > >>>   0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00
> > >>>   0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00
> > >>>   0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00
> > >>>  For input ecut=  1.000000E+01 best grid ngfft=     288     288      24
> > >>>        max ecut=  1.023281E+01
> > >>>  getng: value of mgfft=     288 and nfft=     1990656
> > >>>  getng: values of ngfft(4),ngfft(5),ngfft(6)     289     289      24
> > >>>  getmpw: optimal value of mpw=  110686
> > >>>
> > >>>  iofn2 : deduce lmnmax  =   4, lnmax  =   2,
> > >>>                 lmnmaxso=   4, lnmaxso=   2.
> > >>>  memory : analysis of memory needs
> > >>> ================================================================================
> > >>>  Values of the parameters that define the memory need for DATASET  1.
> > >>>    intxc =         0  ionmov =         0    iscf =         5     ixc =         1
> > >>>   lmnmax =         2   lnmax =         2   mband =       120  mffmem =         1
> > >>> P  mgfft =       288   mkmem =         1 mpssoang=         3     mpw =    110686
> > >>>   mqgrid =      3001   natom =        28    nfft =   1990656    nkpt =         4
> > >>>   nloalg =         4  nspden =         1 nspinor =         1  nsppol =         1
> > >>>     nsym =         1  n1xccc =      2501  ntypat =         2  occopt =         1
> > >>> ================================================================================
> > >>> P This job should need less than                     608.986 Mbytes of memory.
> > >>>   Rough estimation (10% accuracy) of disk space for files :
> > >>>   WF disk file :    810.690 Mbytes ; DEN or POT disk file :     15.190 Mbytes.
> > >>> ================================================================================
> > >>>
> > >>>  Biggest array : cg(disk), with    202.6741 MBytes.
> > >>> -P-0000  leave_test : synchronization done...
> > >>>  memana : allocated an array of    202.674 Mbytes, for testing purposes.
> > >>>  memana : allocated     608.986 Mbytes, for testing purposes.
> > >>>  The job will continue.
> > >>>
> > >>>  DATASET    2 : space group P1 (#  1); Bravais aP (primitive triclinic)
> > >>>  inkpts: Sum of    4 k point weights is    4.000000
> > >>>    With present occopt=   1 , renormalize it to one
> > >>>
> > >>>  inkpts : istwfk preprocessed, gives following first values (max. 6): 1 1 1 1
> > >>>  chkneu : initialized the occupation numbers for occopt=    1
> > >>>     spin-unpolarized case :
> > >>>   2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00
> > >>>   2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00
> > >>>   2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00  2.00
> > >>>   2.00  2.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00
> > >>>   0.00  0.00
> > >>>  For input ecut=  1.000000E+01 best grid ngfft=     288     288      24
> > >>>        max ecut=  1.023281E+01
> > >>>  getng: value of mgfft=     288 and nfft=     1990656
> > >>>  getng: values of ngfft(4),ngfft(5),ngfft(6)     289     289      24
> > >>>  getmpw: optimal value of mpw=  110686
> > >>>
> > >>>  iofn2 : deduce lmnmax  =   4, lnmax  =   2,
> > >>>                 lmnmaxso=   4, lnmaxso=   2.
> > >>>  memory : analysis of memory needs
> > >>> ================================================================================
> > >>>  Values of the parameters that define the memory need for DATASET  2.
> > >>>    intxc =         0  ionmov =         0    iscf =         5     ixc =         1
> > >>>   lmnmax =         2   lnmax =         2   mband =        50  mffmem =         1
> > >>> P  mgfft =       288   mkmem =         1 mpssoang=         3     mpw =    110686
> > >>>   mqgrid =      3001   natom =        28    nfft =   1990656    nkpt =         4
> > >>>   nloalg =         4  nspden =         1 nspinor =         1  nsppol =         1
> > >>>     nsym =         1  n1xccc =      2501  ntypat =         2  occopt =         1
> > >>> ================================================================================
> > >>> P This job should need less than                     490.135 Mbytes of memory.
> > >>>   Rough estimation (10% accuracy) of disk space for files :
> > >>>   WF disk file :    337.789 Mbytes ; DEN or POT disk file :     15.190 Mbytes.
> > >>> ================================================================================
> > >>>
> > >>>  Biggest array : f_fftgr(disk), with    151.8770 MBytes.
> > >>> -P-0000  leave_test : synchronization done...
> > >>>  memana : allocated an array of    151.877 Mbytes, for testing purposes.
> > >>>  memana : allocated     490.135 Mbytes, for testing purposes.
> > >>>  The job will continue.
> > >>>  -outvars: echo values of preprocessed input variables --------
> > >>>      acell    1.0000000000E+02  1.0000000000E+02  7.3174348315E+00 Bohr
> > >>>        amu    2.80855000E+01  1.00794000E+00
> > >>>     diemac    1.20000000E+01
> > >>>       ecut    1.00000000E+01 Hartree
> > >>>    ecuteps1   0.00000000E+00 Hartree
> > >>>    ecuteps2   4.60000000E+00 Hartree
> > >>>    ecutwfn1   0.00000000E+00 Hartree
> > >>>    ecutwfn2   3.10000000E+00 Hartree
> > >>>     getkss1        0
> > >>>     getkss2       50
> > >>>     jdtset      1    2
> > >>>        kpt    0.00000000E+00  0.00000000E+00 -1.25000000E-01
> > >>>               0.00000000E+00  0.00000000E+00 -3.75000000E-01
> > >>>               0.00000000E+00  0.00000000E+00  1.25000000E-01
> > >>>               0.00000000E+00  0.00000000E+00  3.75000000E-01
> > >>> P    mkmem         1
> > >>>      natom        28
> > >>>      nband1      120
> > >>>      nband2       50
> > >>>     ndtset         2
> > >>>      ngfft       288     288      24
> > >>>       nkpt         4
> > >>>      nstep       200
> > >>>       nsym         1
> > >>>     ntypat         2
> > >>>        occ1   2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>        occ2   2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  2.000000  2.000000  2.000000  2.000000
> > >>>               2.000000  2.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000  0.000000  0.000000  0.000000  0.000000
> > >>>               0.000000  0.000000
> > >>>  optdriver1        0
> > >>>  optdriver2        3
> > >>>     ppmfrq1   0.00000000E+00 Hartree
> > >>>     ppmfrq2   6.87212369E-01 Hartree
> > >>>    spgroup         1
> > >>>  symmorphi         0
> > >>>     toldfe    1.00000000E-06 Hartree
> > >>>      typat    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
> > >>>        wtk      0.25000    0.25000    0.25000    0.25000
> > >>>     xangst    2.1774419226E+01  2.6458859148E+01  9.6927622483E-01
> > >>>               ..................
> > >>>               ..................
> > >>>               ..................
> > >>>               ..................
> > >>>      xcart    4.1147688868E+01  4.9999997369E+01  1.8316666044E+00
> > >>>               ..................
> > >>>               ..................
> > >>>               ..................
> > >>>               ..................
> > >>>       xred    4.1147688868E-01  4.9999997369E-01  2.5031539694E-01
> > >>>               ..................
> > >>>               ..................
> > >>>               ..................
> > >>>               ..................
> > >>>      znucl     14.00000    1.00000
> > >>>
> > >>> ================================================================================
> > >>> -P-0000  leave_test : synchronization done...
> > >>>
> > >>>  chkinp: machine precision is   2.2204460492503131E-16
> > >>>
> > >>>  chkinp: Checking input parameters for consistency, jdtset= 1.
> > >>>
> > >>>  chkinp: Checking input parameters for consistency, jdtset= 2.
> > >>> -P-0000
> > >>> -P-0000 ================================================================================
> > >>> -P-0000 == DATASET  1 ==================================================================
> > >>> -P-0000
> > >>>
> > >>>  iofn2 : deduce lmnmax  =   4, lnmax  =   2,
> > >>>                 lmnmaxso=   4, lnmaxso=   2.
> > >>>   distrb2: enter
> > >>>   mpi_enreg%parareel= 0
> > >>>   mpi_enreg%paralbd= 0
> > >>>   mpi_enreg%paral_compil_respfn= 0
> > >>>   distrb2: exit
> > >>>  Unit cell volume ucvol=  7.3174348E+04 bohr3
> > >>>  Angles (23,13,12)=  9.00000000E+01  9.00000000E+01  9.00000000E+01 degrees
> > >>>
> > >>>  getcut: wavevector=  0.0000  0.0000  0.0000  ngfft= 288 288  24
> > >>>          ecut(hartree)=     10.000   => boxcut(ratio)=   2.02315
> > >>> -P-0000  leave_test : synchronization done...
> > >>> kpgio: loop on k-points done in parallel
> > >>> - pspatm: opening atomic psp file    14si.pspnc
> > >>>   Troullier-Martins psp for element  Si        Thu Oct 27 17:31:21 EDT 1994
> > >>>   14.00000   4.00000    940714                znucl, zion, pspdat
> > >>>     1    1    2    2      2001   0.00000      pspcod,pspxc,lmax,lloc,mmax,r2well
> > >>>     0   5.907  14.692    1   2.0872718        l,e99.0,e99.9,nproj,rcpsp
> > >>>    0.00000000   0.00000000   0.00000000   0.00000000     rms, ekb1, ekb2, epsatm
> > >>>     1   2.617   4.181    1   2.0872718        l,e99.0,e99.9,nproj,rcpsp
> > >>>    0.00000000   0.00000000   0.00000000   0.00000000     rms, ekb1, ekb2, epsatm
> > >>>     2   0.000   0.000    0   2.0872718        l,e99.0,e99.9,nproj,rcpsp
> > >>>    0.00000000   0.00000000   0.00000000   0.00000000     rms, ekb1, ekb2, epsatm
> > >>>     1.80626423934776    0.22824404341771    1.17378968127746   rchrg,fchrg,qchrg
> > >>>  pspatm: epsatm=    1.43386982
> > >>>          --- l  ekb(1:nproj) -->
> > >>>              0    3.287949
> > >>>              1    1.849886
> > >>>  pspatm: atomic psp has been read  and splines computed
> > >>>
> > >>> - pspatm: opening atomic psp file    1h.pspnc
> > >>>   Troullier-Martins psp for element  H         Thu Oct 27 17:28:54 EDT 1994
> > >>>    1.00000   1.00000    940714                znucl, zion, pspdat
> > >>>     1    1    0    0      2001   0.00000      pspcod,pspxc,lmax,lloc,mmax,r2well
> > >>>     0   7.740  11.990    0   1.5855604        l,e99.0,e99.9,nproj,rcpsp
> > >>>    0.00000000   0.00000000   0.00000000   0.00000000     rms, ekb1, ekb2, epsatm
> > >>>     0.00000000000000    0.00000000000000    0.00000000000000   rchrg,fchrg,qchrg
> > >>>  Note: local psp for atom with Z=  1.0
> > >>>  pspatm: epsatm=    0.04198703
> > >>>          --- l  ekb(1:nproj) -->
> > >>>  pspatm: atomic psp has been read  and splines computed
> > >>>
> > >>>    1.78187787E+03                                ecore*ucvol(ha*bohr**3)
> > >>> -P-0000  wfconv:   120 bands initialized randomly with npw=110435, for ikpt=     1
> > >>> -P-0000  leave_test : synchronization done...
> > >>>  newkpt: loop on k-points done in parallel
> > >>>  pareigocc : MPI_ALLREDUCE
> > >>>
> > >>>  setup2: Arith. and geom. avg. npw (full set) are  110560.500  110560.429
> > >>>  enter initro
> > >>>  initro : for itypat=  1, take decay length=      1.1000,
> > >>>  initro : indeed, coreel=     10.0000, nval=  4 and densty=  0.0000E+00.
> > >>>  initro : for itypat=  2, take decay length=      0.6000,
> > >>>  initro : indeed, coreel=      0.0000, nval=  1 and densty=  0.0000E+00.
> > >>>
> > >>> ================================================================================
> > >>>
> > >>>      iter   Etot(hartree)      deltaE(h)  residm     vres2    diffor    maxfor
> > >>>
> > >>>  getcut: wavevector=  0.0000  0.0000  0.0000  ngfft= 288 288  24
> > >>>          ecut(hartree)=     10.000   => boxcut(ratio)=   2.02315
> > >>>
> > >>>  ewald : nr and ng are    8 and   45
> > >>>
> > >>>  ITER STEP NUMBER     1
> > >>>  vtorho : nnsclo_now=  2, note that nnsclo,dbl_nnsclo,istep=  0 0  1
> > >>> -P-0000  leave_test : synchronization done...
> > >>>  vtorho: loop on k-points and spins done in parallel
> > >>>
> > >>> ,Min el dens=  8.7518E-07 el/bohr3 at reduced coord.  0.0069  0.1493  0.4583
> > >>> ,Max el dens=  2.1994E-01 el/bohr3 at reduced coord.  0.6042  0.5000  0.4583
> > >>>  ETOT  1  -77.525767923602    -7.753E+01 1.708E-02 1.241E+03 3.756E-02 3.756E-02
> > >>>  scprqt: <Vxc>= -3.1920845E-02 hartree
> > >>>  scfcge:
> > >>>  scfcge:istep-iline_cge-ilinmin lambda      etot             resid
> > >>>  scfcge: actual     1-0-0   0.0000E+00 -7.752576792360E+01  4.8328E+01
> > >>>
> > >>>   .............................................
> > >>>   .............................................
> > >>>
> > >>>  ITER STEP NUMBER    22
> > >>>  vtorho : nnsclo_now=  1, note that nnsclo,dbl_nnsclo,istep=  0 0 22
> > >>> -P-0000  leave_test : synchronization done...
> > >>>  vtorho: loop on k-points and spins done in parallel
> > >>>
> > >>> ,Min el dens=  2.7450E-10 el/bohr3 at reduced coord.  0.9410  0.0417  0.2500
> > >>> ,Max el dens=  2.1254E-01 el/bohr3 at reduced coord.  0.6042  0.5000  0.4583
> > >>>  ETOT 22  -77.731665580830    -5.966E-08 3.688E-06 4.358E-02 4.241E-05 4.895E-04
> > >>>  scprqt: <Vxc>= -1.7237925E-02 hartree
> > >>>
> > >>>  At SCF step   22, etot is converged :
> > >>>   for the second time, diff in etot=  5.966E-08 < toldfe=  1.000E-06
> > >>> -P-0000  leave_test : synchronization done...
> > >>> forstrnps: loop on k-points and spins done in parallel
> > >>> -P-0000  leave_test : synchronization done...
> > >>> forstrnps: loop on k-points and spins done in parallel
> > >>>
> > >>>  Cartesian components of stress tensor (hartree/bohr3)
> > >>>   sigma(1 1)=  1.63063121E-06  sigma(3 2)= -1.93262486E-12
> > >>>   sigma(2 2)=  1.70719506E-06  sigma(3 1)=  1.34868679E-08
> > >>>   sigma(3 3)=  1.47931180E-06  sigma(2 1)= -2.95973572E-10
> > >>>
> > >>>
> > >>> ======================================================================
> > >>>  Calculating and writing out Kohn-Sham electronic Structure file
> > >>>  Using diagonalized wavefunctions and energies (kssform=1)
> > >>>
> > >>>  Additional amount of memory required by "outkss" routine= 1976.64 Mbytes.
> > >>>
> > >>>  Sorting g-vecs for an output of states on an unique "big" PW basis.
> > >>>  The inversion was not found in the symmetries list.
> > >>>  - outkss - COMMENT :
> > >>>    GW program uses the original set of symmetries
> > >>>  Shells found:
> > >>>  number of shell    number of G vectors      cut-off energy
> > >>>                1                      1               0.000
> > >>>                2                      2               0.002
> > >>>                3                      3               0.002
> > >>>                4                      4               0.002
> > >>>                5                      5               0.002
> > >>>                6                      6               0.004
> > >>>                7                      7               0.004
> > >>>                8                      8               0.004
> > >>>                9                      9               0.004
> > >>>               10                     10               0.008
> > >>>               11                     11               0.008
> > >>>               12                     12               0.008
> > >>>               13                     13               0.008
> > >>>               14                     14               0.010
> > >>>               15                     15               0.010
> > >>>               16                     16               0.010
> > >>>               17                     17               0.010
> > >>>               18                     18               0.010
> > >>>               19                     19               0.010
> > >>>               20                     20               0.010
> > >>>               21                     21               0.010
> > >>>               22                     22               0.016
> > >>>               23                     23               0.016
> > >>>               24                     24               0.016
> > >>>               25                     25               0.016
> > >>>               26                     26               0.018
> > >>>               27                     27               0.018
> > >>>               28                     28               0.018
> > >>>               29                     29               0.018
> > >>>               30                     30               0.020
> > >>>               31                     31               0.020
> > >>>               32                     32               0.020
> > >>>               33                     33               0.020
> > >>>               34                     34               0.020
> > >>>               35                     35               0.020
> > >>>               36                     36               0.020
> > >>> ...............................
> > >>>
> > >>>             ****                   ****              11.326
> > >>>             ****                   ****              11.326
> > >>>             ****                   ****              11.326
> > >>>             ****                   ****              11.326
> > >>>
> > >>>
> > >>>  Since the number of g's to be written on file
> > >>>  was 0 or too large, it has been set to the max. value.,
> > >>>  computed from the union of the sets of G vectors for the different k-points.
> > >>>  Number of g-vectors written on file is: *****
> > >>>  Number of bands written on file is:    50
> > >>>
> > >>>  Since the number of bands to be computed
> > >>>  is less than the number of G-vectors found,
> > >>>  the program will perform partial diagonalizations.
> > >>>
> > >>>  Amount of disk space required by _STA file=  378.38 Mbytes.
> > >>>
> > >>>  Opening file for KS structure output: Sio_DS1_KSS
> > >>>  number of Gamma centered plane waves 122911
> > >>>  number of Gamma centered shells 122911
> > >>>  number of bands     50
> > >>>  maximum angular momentum components      3
> > >>>  number of symmetry operations  1 (without inversion)
> > >>>    *   g( 1)=  0  0  0
> > >>>    *   g( 2)=  1  0  0
> > >>>    *   g( 3)= -1  0  0
> > >>>    *   g( 4)=  0  1  0
> > >>>    *   g( 5)=  0 -1  0
> > >>>    *   g( 6)=  1  1  0
> > >>>    *   g( 7)= -1  1  0
> > >>>    *   g( 8)=  1 -1  0
> > >>>    *   g(****)=-30 13 -5
> > >>>    *   g(****)= 13 30 -5
> > >>>    *   g(****)=-13 30 -5
> > >>>    *   g(****)= 13-30 -5
> > >>>    *   g(****)=-13-30 -5
> > >>>    *   g(****)= 30-13 -5
> > >>>    *   g(****)=-30-13 -5
> > >>> -P-0000
> > >>> -P-0000  k-point   1
> > >>> -P-0000  Calculating <G|H|G'> elements
> > >>>
          
> > >>
> > >>
        
> > > -- 
> > > Dr. Riad Shaltaf                   UCL/SE/FSA/MAPR/PCPM
> > > Tel: +32 (0)10 47 24 51            Bâtiment Boltzmann, a+1
> > > Fax: +32 (0)10 47 34 52            1 place Croix du Sud
> > > Mel: shaltaf@pcpm.ucl.ac.be        1348 Louvain-la-Neuve (Belgique)
> > >
> > >
      
-- 
***************************************************************************
Deyu Lu (Ph.D)

190 Chemistry Building
University of California, Davis 
One Shields Avenue
Davis, CA 95616
Office phone: (530) 754-9663
Group Webpage: http://angstrom.ucdavis.edu/

***************************************************************************


ธนูสิทธิ์ บุรินทร์ประโคน wrote:
NextMessageA99408FD-62AC-10FA-0A01-F1297F16FD4At@mail.kku.ac.th" type="cite"> 
Dear All

I've tried to perform the GW calculations for GaP band structue using 
parallelized Abinit 5.4.4(MPI version, prepared for a x86_64_linux) but
encounter a problem that the program crashed during writing out the KSS
file. The error message at the end of the log file is as follows.(Please
see the complete "log" and "output" files attached. I'd like to keep
this mail clear.)

..............................................................

-P-0000
-P-0000  k-point   9
 Eigenvalues in Hartrees for ikpt=  9:
   9     -0.4690 -0.4646 -0.4645 -0.4629 -0.4616 -0.3341 -0.0907 -0.0023
 0.0046
          0.1554  0.2057  0.3288  0.3456  0.4092  0.4942  0.5610  0.6209
 0.6601
          0.6900  0.7058  0.7164  0.8253  0.8639  0.9599  1.0117  1.0742
 1.1571
          1.1582  1.1892  1.1934  1.3033  1.3345  1.3405  1.4062  1.4376
 1.5037
          1.5377  1.5465  1.5719  1.5827  1.6130  1.6472  1.6823  1.7137
 1.7426
          1.7541  1.8023  1.8373  1.8965  1.9004  1.9265  2.0175  2.0283
 2.0539
          2.0606  2.1634  2.1777  2.1954  2.2432  2.2616  2.2982  2.3398
 2.3649
          2.3907  2.4075  2.5007  2.5362  2.5646  2.5932  2.6236  2.6375
 2.6395
          2.6635  2.6873  2.6988  2.7386  2.7775  2.7811  2.8197  2.8319
 2.8674
rank 0 in job 3  localhost.localdomain_41152   caused collective abort
of all ranks
  exit status of rank 0: killed by signal 11 

..........................................................


My platform is Fedora 8.0 running on an Intel Core 2 Quad CPU with 4 Gb
of RAM. The above run was carried out with 4 processors. 

I firstly carried out the usual ground state density calculation and
then used it as the input data for the KSS calculation routine. I either
ran the two calculations in-one-go or separated them into different runs
to make sure that the RAM was sufficient(maybe a bad guess) for the KSS
calculation. The ground state calculation was completed nicely. However,
in the KSS calculation, it seems from the output and log files that the
non-SCF
calculation has been completed for all k-points but the program
terminated during writing out the KSS file the Eigenvalues for ikpt =9,
as shown above.

What might be the possible causes for this?
Is it related to memory issue?

Could somebody help me on this, please?. Thank you very much.
 

My input files are listed below.



Kind regards,
Thanusit Burinprakhon


Input files:

# Crystalline GaP
# Preparation KSS file, including a Gamma point, for convergence test at 
# Gamma 

.........................................
#Input used to calculate GS density

#Dataset 1 : usual self-consistent calculation for ground state charge 
#density
#kptopt1 1          # Option for the automatic generation of k points,
                   # taking into account the symmetry
#nshiftk1 4
#shiftk1  0.5 0.5 0.5  # These shifts will be the same for all grids
#         0.5 0.0 0.0
#         0.0 0.5 0.0
#         0.0 0.0 0.5
#ngkpt1  6 6 6         # Optimum k-grid for Etot convergence
#nband1  15
#prtden1  1            # Print the density, for use by dataset 2
#iscf1 5
#toldfe1  1.0d-8
........................................

# calculation of kss file

## Definition of k-points
kptopt 0      # Option for manually provided kpts.
nkpt   29     # Gamma point and those similar to  
                   # kpts used in the GS calculation 
kpt     0.0 0.0 0.0      
       -8.33333333E-02 -1.66666667E-01  0.00000000E+00
       -8.33333333E-02 -3.33333333E-01  0.00000000E+00
       -1.66666667E-01 -2.50000000E-01  0.00000000E+00
       -8.33333333E-02 -2.50000000E-01  8.33333333E-02
       -8.33333333E-02  5.00000000E-01  0.00000000E+00
       -1.66666667E-01 -4.16666667E-01  0.00000000E+00
       -8.33333333E-02 -4.16666667E-01  8.33333333E-02
       -2.50000000E-01 -3.33333333E-01  0.00000000E+00
       -1.66666667E-01 -3.33333333E-01  8.33333333E-02
       -8.33333333E-02 -3.33333333E-01  1.66666667E-01
       -8.33333333E-02  3.33333333E-01  0.00000000E+00
       -1.66666667E-01  4.16666667E-01  0.00000000E+00
       -2.50000000E-01  5.00000000E-01  0.00000000E+00
       -1.66666667E-01  5.00000000E-01  8.33333333E-02
       -3.33333333E-01 -4.16666667E-01  0.00000000E+00
       -2.50000000E-01 -4.16666667E-01  8.33333333E-02
       -1.66666667E-01 -4.16666667E-01  1.66666667E-01
       -8.33333333E-02 -4.16666667E-01  2.50000000E-01
       -8.33333333E-02  1.66666667E-01  0.00000000E+00
       -1.66666667E-01  2.50000000E-01  0.00000000E+00
       -2.50000000E-01  3.33333333E-01  0.00000000E+00
       -3.33333333E-01  4.16666667E-01  0.00000000E+00
       -4.16666667E-01  5.00000000E-01  0.00000000E+00
       -3.33333333E-01  5.00000000E-01  8.33333333E-02
       -2.50000000E-01  5.00000000E-01  1.66666667E-01
       -8.33333333E-02  0.00000000E+00  0.00000000E+00
       -2.50000000E-01  0.00000000E+00  0.00000000E+00
       -4.16666667E-01  0.00000000E+00  0.00000000E+00 

istwfk  29*1           
iscf     -2             # Non self-consistent calculation
getden    1          # Read previously calculated density file
tolwfr   1.0d-10     
nband    310
nbandkss 300      # Number of bands in KSS file
nbdbuf    10
kssform   3

#Common input parameters

ecut  100 
acell  3*10.2343       
rprim  0.0  0.5  0.5    
          0.5  0.0  0.5   
          0.5  0.5  0.0
ntypat 2          
znucl 15 31       
natom 2           # There are two atoms.
typat 1 2          # 1=P 2=Ga
xred              
     0.0000000000E+00  0.0000000000E+00  0.0000000000E+00
     2.5000000000E-01  2.5000000000E-01  2.5000000000E-01 

nstep   100       # Maximal number of SCF cycles
diemac  12.0      
nsym 0        
symmorphi 0      
ixc  7
prtvol 1 

-- 
D. R. Hamann
Mat-Sim Research LLC    | Deptartment of Physics
P.O. Box 742            |  and Astronomy
Murray Hill, NJ 07974   | Rutgers University
phone: 908-370-8079     | 732-445-4381

email: drhamann@mat-simresearch.com

Archive powered by MHonArc 2.6.16.

Top of Page