[CP2K:4583] error: cp_fm_cholesky_decompose processor

[marcella Iannuzzi]

Hi, 

The error is caused by an ill conditioned overlap matrix.
This generates numerical problems  in the construction of the preconditioner, when the cholesky decomposition of the overlap matrix is needed.
You can choose a method that does not need the decomposition of the overlap matrix (e.g. another preconditioner).
Maybe it is enough to do so for the first steps of the optimization, and then switch back.
You can also try to slightly displace some atoms, in order to break symmetries. 
However, the best solution would be to change basis sets. The basis sets in the  BASIS_MOLOPT library should be safer in this respect. 

kind regards
 Marcella


--------------------------------------------------------------
Marcella Iannuzzi              Phone : ++41 44 635 4479
Physical Chemistry Institute   FAX   : ++41 44 635 6838
University of Zurich               E-mail:  marc... at pci.uzh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
---------------------------------------------------------------




On Aug 26, 2013, at 3:52 PM, anirban mondal wrote:

> Dear CP2K users/developers,
>               I am trying to optimize a system (periodic) using the aug-TZV2P-GTH basis set. I have got an error, appended following.
> 
> **********************************************
>  Step     Update method      Time    Convergence         Total energy    Change
>   ------------------------------------------------------------------------------
> 
>   Trace(PS):                                  184.0000000000
>   Electronic density on regular grids:       -183.9999999144        0.0000000856
>   Core density on regular grids:              183.9999999874       -0.0000000126
>   Total charge density on r-space grids:        0.0000000730
>   Total charge density g-space grids:           0.0000000730
> 
>  cholesky_decomposition error at vector          450
> 
>  *****************************************************************************
>  *** 19:10:38 ERRORL2 in cp_fm_cholesky:cp_fm_cholesky_decompose processor ***
>  *** 0  err=-300 condition FAILED at line 98                               ***
>  *****************************************************************************
> 
> 
>  ===== Routine Calling Stack ===== 
> 
>             8 cp_fm_cholesky_decompose
>             7 make_full_all
>             6 make_preconditioner
>             5 prepare_preconditioner
>             4 init_scf_loop
>             3 scf_env_do_scf
>             2 qs_energies_scf
>             1 CP2K
>  cholesky_decomposition error at vector          450
>  cholesky_decomposition error at vector          450
>  CP2K| condition FAILED at line 98
>  CP2K| Abnormal program termination, stopped by process number 0
>  cholesky_decomposition error at vector          450
>  cholesky_decomposition error at vector          450
>  cholesky_decomposition error at vector          450
> 
> *******************************************************
> 
> I have tried with the keyword "CHOLESKY OFF" in the &SCF section and also played with the "EPS_EIGVAL" option in both the sides (by increasing and decreasing). But it didn't solve the problem.
> 
> INPUT FILE
> *************************************************
> &GLOBAL
>   PROJECT      mmim-cl
>   RUN_TYPE     GEO_OPT
>   PRINT_LEVEL  MEDIUM 
> &END GLOBAL
> 
> &FORCE_EVAL
>   METHOD  Quickstep
> 
>   &DFT
>     BASIS_SET_FILE_NAME  /home1/bala/cp2k/tests/QS/GTH_BASIS_SETS 
>     POTENTIAL_FILE_NAME  /home1/bala/cp2k/tests/QS/POTENTIAL
>     &MGRID
>       CUTOFF    600
>       NGRIDS	5
>       REL_CUTOFF 40 
>     &END MGRID
>     &QS
>       METHOD       GPW    
>       EPS_DEFAULT  1.0E-10 
>     &END QS
> 
>     &SCF
>  #  CHOLESKY OFF
> #   EPS_EIGVAL 1.0
>       MAX_SCF   30 
>       EPS_SCF   1.0E-7
>       SCF_GUESS RESTART
>       &OT ON
>         MINIMIZER       CG
>         PRECONDITIONER  FULL_ALL
>         ENERGY_GAP      0.001
>         STEPSIZE 0.05
>       &END OT
>       &OUTER_SCF
>         MAX_SCF  500
>         EPS_SCF  1.0E-7  
>       &END OUTER_SCF
>     &END SCF
>     &XC
>       &XC_FUNCTIONAL  PBE
>       &END XC_FUNCTIONAL
>       &vdW_POTENTIAL
>          DISPERSION_FUNCTIONAL PAIR_POTENTIAL
>          &PAIR_POTENTIAL
>             TYPE DFTD2
>             REFERENCE_FUNCTIONAL PBE 
>          &END PAIR_POTENTIAL
>       &END vdW_POTENTIAL
>       &XC_GRID
>         XC_DERIV  SPLINE2
>         XC_SMOOTH_RHO NN50
>      &END XC_GRID
>     &END XC
>   &END DFT
> 
>   &SUBSYS
>     &CELL 
>       ABC 8.652 7.858 10.539 
>       ALPHA_BETA_GAMMA  90.0 106.34 90.00
>     &END CELL
>     &COORD
>       @INCLUDE coor.xyz  
>     &END COORD
> 
>     &KIND H
>       BASIS_SET  aug-DZVP-GTH
>       POTENTIAL  GTH-PBE-q1
>     &END KIND 
>     &KIND C
>       BASIS_SET  aug-DZVP-GTH
>       POTENTIAL  GTH-PBE-q4
>     &END KIND 
>     &KIND N
>       BASIS_SET  aug-DZVP-GTH
>       POTENTIAL  GTH-PBE-q5
>     &END KIND 
>     &KIND Cl
>       BASIS_SET  aug-DZVP-GTH
>       POTENTIAL  GTH-PBE-q7 
>     &END KIND
>   &END SUBSYS
> 
> &END FORCE_EVAL
> 
> &MOTION
>   &GEO_OPT
>     MINIMIZER  BFGS 
>     MAX_ITER   50000
>   &END GEO_OPT
> &END MOTION
> 
> 
> *************************************************************
> 
> Any one can give help?
> 
> 
> Thank you very much.
> 
> regards;
> anirban
> 
> 
> 
> -- 
> Anirban Mondal
> Molecular Simulations Lab.
> Jawaharlal Nehru Centre for Advanced Scientific Research
> Bangalore-560064
> India
> 
> -- 
> You received this message because you are subscribed to the Google Groups "cp2k" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns... at googlegroups.com.
> To post to this group, send email to cp... at googlegroups.com.
> Visit this group at http://groups.google.com/group/cp2k.
> For more options, visit https://groups.google.com/groups/opt_out.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20130826/53519f80/attachment.htm>