[CPMD-list] Problems of Kohn-Sham Energy

[Axel Kohlmeyer]

On Mon, 18 Jun 2007, Hua Bao wrote:

HB> Dear CPMD-user,
HB> 
HB> I am using CPMD 3.11.1 and trying to calculate KOHN-SHAM ENERGY of Al2O3.
HB> CPMD is compiled IFORT-AMD64-MPI. It does not have any problem during
HB> geometry optimization. However, when I calculate Kohn-Sham Energy, it shows
HB> the following error message,
HB> .....
HB>  GENERATE ATOMIC BASIS SET
HB>      Al        SLATER ORBITALS
HB>         3S        ALPHA=   1.3724      OCCUPATION= 2.00
HB>         3P        ALPHA=   1.3552      OCCUPATION= 1.00
HB>       O        SLATER ORBITALS
HB>         2S        ALPHA=   2.2458      OCCUPATION= 2.00
HB>         2P        ALPHA=   2.2266      OCCUPATION= 4.00
HB>  
HB> 
HB>  INITIALIZATION TIME:                                0.72 SECONDS
HB> 
HB>  ***    WFOPTS| SIZE OF THE PROGRAM IS   46924/ 875704 kBYTES ***
HB>  RV30! NUMBER OF STATES   HAS CHANGED                     72   82
HB> p2_12560:  p4_error: interrupt SIGSEGV: 11
HB> p1_18787:  p4_error: interrupt SIGSEGV: 11
HB> rm_l_1_18788: (2.097656) net_send: could not write to fd=5, errno = 32
HB> rm_l_2_12561: (1.828125) net_send: could not write to fd=5, errno = 32
HB> p2_12560: (1.828125) net_send: could not write to fd=5, errno = 32
HB> p1_18787: (6.117188) net_send: could not write to fd=5, errno = 32
HB> 
HB> ......

this part of the code has not been touched for years
and most likely, parallel support is broken (if it 
ever worked). please try running in serial. 

if you want to use CPMD seriously for calculations like 
this, particularly on larger systems, i highly recommend, 
that you familiarize yourself with the inner workings of 
the code and get ready to sort out problems by changing 
the sources. the last time that the k-point/bands code had
been significantly updated was over 3 years ago.

if there is anybody out there, who has made improvements
to the k-points code, that are not yet in the latest release,
please speak up so that we avoid having people doing the
same work twice.

thanks,
  axel.

HB> 
HB> Here is my input,
HB> 
HB> &CPMD
HB>  KOHN-SHAM ENERGIES
HB>  10
HB>  RESTART WAVEFUNCTION COORDINATES
HB>  ENERGYBANDS
HB> &END
HB> 
HB> &SYSTEM
HB>  SYMMETRY
HB>  4
HB>  ANGSTROM
HB>  CELL
HB>  4.7602 1 2.7296 0 0 -0.5
HB>  CUTOFF
HB>  60.0
HB>  KPOINTS BANDS block =10
HB>  21 0 0 0            0.5 0 0      gamma to m
HB>  


HB>  0  0 0 0       0 0 0
HB> &END
HB> 
HB> &DFT
HB>  FUNCTIONAL LDA
HB> &END
HB> 
HB> &ATOMS
HB> *Al_SGS_LDA.psp  KLEINMAN-BYLANDER
HB> LMAX=P
HB> 12
HB> 0.0000  0.0000  4.5757
HB> 


HB> 
HB> *O_MT_LDA.psp KLEINMAN-BYLANDER
HB> LMAX=P
HB> 18
HB> 


HB> &END
HB> 
HB> Thanks,
HB> Hua Bao
HB>  
HB> Graduate Student
HB> School of Mechanical Engineering 
HB> Purdue University
HB> Office Tel: (765) 494-6944 
HB> 
HB> 
HB> _______________________________________________
HB> CPMD-list mailing list
HB> CPMD-list at cpmd.org
HB> http://cpmd.org/mailman/listinfo/cpmd-list
HB> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.