[Pw_forum] Parallel version and NFS
Huiqun Zhou
hqzhou at nju.edu.cn
Thu Apr 6 08:12:37 CEST 2006
For convergency problem, you can try to set mixing_beta to a smaller value,
say 0.2.
Then you need check if your ecutwfc and ecutrho need to be set to higher
values.
Huiqun Zhou
----- Original Message -----
From: <sir_puding at tut.by>
To: "Paolo Giannozzi" <pw_forum at pwscf.org>
Sent: Wednesday, April 05, 2006 7:54 PM
Subject: Re[2]: [Pw_forum] Parallel version and NFS
>> I cannot believe that your system has no local scratch filesystem!
> It has, But there is no guarantee that local /tmp has enough space on
> every node. So, i need to use my /home/lds HDD. (Now I'm using simple
> test-cluster which is used by many people before extensive calculations
> on serious machine)
>
>> please provide the input that produces this problem
> I thought that there is no any difference between 'from scratch' and
> 'restart' modes when there is no any files from previous calculation.
> In output program writes 'save not found, using input configuration',
> same for potential and wavefunctions.
> For me it means that calculation goes 'from_scratch'.
>
> Anyway, I've started calculations right, 'from_scratch', with
> potential from atoms superposition and the same for wfc, and using
> local temp dir, but problem with convergence in first iteration still
> exists.
>
> Sergey
>
> my old input
> --------
> &CONTROL
> calculation = 'vc-relax' ,
> restart_mode = 'restart' ,
> wf_collect = .true. ,
> outdir = '/home/lds/tmp/' ,
> pseudo_dir = '/home/lds/espresso/pseudo/' ,
> prefix = 'silicon' ,
> disk_io = 'default' ,
> verbosity = 'default' ,
> etot_conv_thr = 1.D-6 ,
> forc_conv_thr = 1.D-6 ,
> nstep = 30 ,
> tstress = .true. ,
> tprnfor = .true. ,
> dt = 50 ,
> /
> &SYSTEM
> ibrav = 0,
> celldm(1) = 13.4834,
> nat = 64,
> ntyp = 2,
> ecutwfc = 30 ,
> nosym = .true. ,
> nbnd = 300,
> nelec = 257,
> occupations = 'tetrahedra' ,
> nspin = 1 ,
> /
> &ELECTRONS
> conv_thr = 1.0d-8 ,
> startingpot = 'file' ,
> startingwfc = 'file' ,
> mixing_mode = 'plain' ,
> mixing_beta = 0.7 ,
> diagonalization = 'david' ,
> /
> &IONS
> ion_dynamics = 'damp' ,
> /
> &CELL
> cell_dynamics = 'damp-pr' ,
> /
> CELL_PARAMETERS alat
> 1.001284827 0.002853981 0.002846682
> 0.002853985 1.001308791 0.002853985
> 0.002846682 0.002853981 1.001284827
> ATOMIC_SPECIES
> C 12.01078 C.pbe-rrkjus.UPF
> N 14.00675 N.pbe-rrkjus.UPF
> ATOMIC_POSITIONS alat
> C -0.000411174 -0.000204767 -0.000411174
> C 0.002098331 0.002221908 0.500932170
> C -0.000045859 0.250580894 0.250457544
> C 0.002219314 0.252923640 0.750880737
> C 0.002109330 0.501077678 0.002109331
> C 0.001767192 0.501795146 0.501650854
> C 0.002186471 0.751016286 0.252798804
> C 0.005447910 0.753580185 0.753419772
> C 0.125108721 0.125251176 0.125108721
> C 0.126543635 0.126666058 0.625869486
> C 0.125324871 0.375989285 0.375861475
> C 0.128972956 0.378221350 0.876231944
> C 0.126537116 0.626028922 0.126537116
> C 0.128722812 0.627568057 0.627379082
> C 0.128970236 0.876359204 0.378078334
> C 0.130887959 0.878902709 0.878695675
> C 0.250439611 0.000061706 0.250439611
> C 0.252807018 0.002343198 0.750849658
> C 0.250457543 0.250580894 -0.000045859
> C 0.249832391 0.249950783 0.500836235
> C 0.249831988 0.500975286 0.249831988
> C 0.255023896 0.504124094 0.750606158
> C 0.252798803 0.751016286 0.002186471
> C 0.255037561 0.750757592 0.503984572
> C 0.375856435 0.125431490 0.375856435
> C 0.378088203 0.129097780 0.876235649
> C 0.375861474 0.375989285 0.125324871
> C 0.375437799 0.375557771 0.622013549
> C 0.375445558 0.622149847 0.375445558
> C 0.380886166 0.629192100 0.877632827
> C 0.378078334 0.876359204 0.128970236
> C 0.380897096 0.877778485 0.629053129
> C 0.500932169 0.002221908 0.002098332
> C 0.501653977 0.001878686 0.501653977
> C 0.500836235 0.249950783 0.249832392
> C 0.503997539 0.255156665 0.750631256
> C 0.501650854 0.501795146 0.001767192
> N 0.486107094 0.486200194 0.486107094
> C 0.503984572 0.750757592 0.255037561
> C 0.508233521 0.756508940 0.756374101
> C 0.625869485 0.126666058 0.126543636
> C 0.627425583 0.128878572 0.627425584
> C 0.622013549 0.375557771 0.375437800
> C 0.629059253 0.381018969 0.877650922
> C 0.627379082 0.627568058 0.128722812
> C 0.648656738 0.648807613 0.648656738
> C 0.629053128 0.877778485 0.380897097
> C 0.630868728 0.882939575 0.882796189
> C 0.750849657 0.002343198 0.252807018
> C 0.753422309 0.005560609 0.753422310
> C 0.750880737 0.252923640 0.002219315
> C 0.750631256 0.255156665 0.503997539
> C 0.750606157 0.504124094 0.255023897
> C 0.756381813 0.508368160 0.756381814
> C 0.753419771 0.753580185 0.005447910
> C 0.756374101 0.756508940 0.508233522
> C 0.876235649 0.129097780 0.378088203
> C 0.878717335 0.130994359 0.878717336
> C 0.876231943 0.378221351 0.128972956
> C 0.877650922 0.381018969 0.629059253
> C 0.877632827 0.629192100 0.380886166
> C 0.882799518 0.631017191 0.882799519
> C 0.878695675 0.878902709 0.130887959
> C 0.882796188 0.882939575 0.630868729
> K_POINTS automatic
> 4 4 4 1 1 1
> ---------------
> --
> Please avoid sending me Word or PowerPoint attachments.
> See http://www.gnu.org/philosophy/no-word-attachments.html
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
>
More information about the Pw_forum
mailing list