The ABINIT Users Mailing List ( CLOSED )

[Fabien Bruneval <fabien.bruneval@cea.fr>]

Dear Tieyu,
I guess you are running out of memory.
Your system has 60000 PW and as you use the default "kssform 1", Abinit 
tries to allocate a 60000x60000 Hamiltonian matrix...

Try to learn how to use "kssform 3" that avoids the full diagonalization 
of the hamiltonian.

Fabien



mailoliver wrote:
> Dear Fabien:
> I use abinis to calculate beta-SiC 64 atoms supercell with cubic 
> structure and only one kpoint(0 0 0).But it still could not write 
> complete KSS file. And there are message:
> forrtl: severe (174): SIGSEGV, segmentation fault occurred
> Image PC Routine Line Source
> abinis 0816FACF Unknown Unknown Unknown
> abinis 0812C955 Unknown Unknown Unknown
> abinis 080A3236 Unknown Unknown Unknown
> abinis 0805E230 Unknown Unknown Unknown
> abinis 08052E51 Unknown Unknown Unknown
> abinis 0804BA45 Unknown Unknown Unknown
> abinis 08049A4A Unknown Unknown Unknown
> libc.so.6 555F0E80 Unknown Unknown Unknown
> abinis 08049971 Unknown Unknown Unknown
> Input and log files are in attachment. Could you give me more advice?
> Best
> Tieyu
>
>
> ?2008-10-05,"BRUNEVAL Fabien 201177" <fabien.bruneval@cea.fr> ??:
> >Dear Tieyu,
> >
> >Good that you could reproduce the published numbers!
> >You should know that, by now, the production of KSS files is not compatible 
> with the parallelism over bands and FFTs.
> >In principle, you can generate the KSS file using only the "old" 
> parallelisation over k-points. So I would advise to remove from your input file all the 
> keywords related to fft_opt_lob, nbdblock etc...
> >
> >By the way, your k-point grid is not regular? or your supercell is not cubic?
> >
> >Bye.
> >
> >Fabien
> >
> >
> >
> >-------- Message d'origine--------
> >De: mailoliver@163.com [mailto:mailoliver@163.com]
> >Date: dim. 10/5/2008 11:15
> >À: forum@abinit.org
> >Objet : [abinit-forum] parallel calculation for GWA
> > 
> >Dear all:
> >
> >I try to repeat the GWA results of beta-SiC in PRB 78, 085125 (2008). I can 
> get
> >the similar results for the system, which include only two atoms. But when I
> >try to run supercell with 64 atoms, I could not get KSS file. I find the
> >processes are static.  Is the file too large, and ifort could not write it? 
> Or
> >maybe I make som mistakes.
> >
> >Best,
> >
> >
> >Tieyu  
> >
> >I compile abinit-5.6.0 with mpich-ifort.9.0, and run abinip with 8 
> processors.
> >
> >
> >my input file:
> >*******************************************
> >ndtset         3
> >
> ># Dataset1: usual self-consistent ground-state calculation
> ># Definition of the k-point grid
> >kptopt1  1        # Option for the automatic generation of k points,
> >ngkpt1       8 8 1
> >nshiftk1 1
> >shiftk1  0. 0. 0.  # These shifts will be the same for all grids
> ># Definition of the SCF procedure
> >toldfe1  1.0d-6    # Will stop when this tolerance is achieved on total 
> energy
> >prtden1  1     # Print out density
> >
> ># Dataset2: calculation of kss file
> ># Definition of k-points
> >kptopt2  0          # K-points will be provided
> >nkpt2        1              # Take only 1 k-point:
> >kpt2         0.0  0.0  0.0  # the Gamma point
> >istwfk2  1          # Option needed for special k-points like Gamma
> ># Definition of the SCF procedure
> >iscf2        -2             # Non self-consistent calculation
> >getden2  1         # Read previous density file
> >tolwfr2  1.0d-16    # Still get it converged
> >nband2       280
> >
> ># Definition of parameters for the calculation of the kss file
> >nbandkss2  480          # Number of bands in KSS file (the maximum possible)
> >wfoptalg2 4
> >nbdblock2 8
> >fft_opt_lob2 2
> >fftalg2 401
> >
> ># Dataset3: Calculation of the screening (epsilon^-1 matrix)
> >optdriver3  3
> >getkss3     -1
> >nband3          480
> >ecutwfn3    3.6
> >ecuteps3    6.0
> >ppmfrq3    16.7 eV
> >gwpara3    2
> >.........
> >
>   
>
>
> ------------------------------------------------------------------------
> [??] ??????26??? <http://popme.163.com/link/004584_1007_8510.html>