[Pw_forum] HSE calculation stops progressing after convergence

Giuseppe Mattioli giuseppe.mattioli at ism.cnr.it
Fri Feb 16 12:46:53 CET 2018


Dear Shoaib Muhammad

First of all EXX calculations are *far more stable* when you use norm  
conserving pseudopotentials. ultrasoft pseudopotentials such as yours  
do not even guarantee particular speedup: the bottleneck is on density  
cutoff rather than on wavefunction cutoff.

Moreover, EXX calculations are generally more expensive than GGA, even  
at Gamma (and your cell is small and would require a proper k-points  
mesh). As you did not post further details, I can only suggest that you:

1) check that the memory requirements for your calculation are  
compatible with your actual resources; an (over)estimate of the total  
memory is printed at the beginning of the output, before the beginning  
of scf iterations.

2) use a recent version of the code (6.2 or newer), which  
automatically implement the ACE algorithm which speedups the calculation

3) play a bit with ecutfock (ecutfock=2*ecutrho is generally safe,  
even less sometimes) which saves a lot of memory

HTH
Giuseppe

Quoting Shoaib <mshoaibce at gmail.com>:

> Dear Members,
>
>
>
> I am trying a simple HSE calculation but it stop progressing after some time
> where output file says "convergence has been achieved in  10 iterations"
> without showing any further error. CPU usage still remains high but nothing
> is further printed in the output file even if I wait for hours. Here is my
> input file:
>
>
>
>
>
> &CONTROL
>
>     calculation = "scf"
>
>     max_seconds =  8.64000e+04
>
>     outdir      = "./"
>
>     prefix      = "espresso"
>
>     pseudo_dir  = " \.pseudopot"
>
>     title       = "LiFePO4 (SCF)"
>
>     wf_collect  = .TRUE.
>
>     wfcdir      = "./"
>
>     tstress     = .TRUE.
>
>     tprnfor     = .true.
>
>     verbosity   = 'high'
>
> /
>
>
>
> &SYSTEM
>
>     a                         =  4.75196e+00
>
>     b                         =  6.08735e+00
>
>     c                         =  1.04507e+01
>
>     degauss                   =  1.00000e-02
>
>     ecutrho                   =  3.50000e+02
>
>     ecutwfc                   =  3.50000e+01
>
>     exxdiv_treatment          = "gygi-baldereschi"
>
>     ibrav                     = 8
>
>     input_dft                 = "hse"
>
>     nat                       = 28
>
>     nbnd                      = 132
>
>     nqx1                      = 1
>
>     nqx2                      = 1
>
>     nqx3                      = 1
>
>     exx_fraction              = 0.25
>
>     ntyp                      = 4
>
>     occupations               = "smearing"
>
>     smearing                  = "gaussian"
>
>     x_gamma_extrapolation     = .TRUE.
>
> /
>
>
>
> &ELECTRONS
>
>     conv_thr         =  1.00000e-06
>
>     electron_maxstep = 200
>
>     adaptive_thr     = .TRUE.
>
>     mixing_beta      =  4.00000e-01
>
>     startingpot      = "atomic"
>
>     startingwfc      = "atomic+random"
>
> /
>
>
>
> K_POINTS {gamma}
>
> ATOMIC_SPECIES
>
> Li      6.94100  Li.pbe-n-van.UPF
>
> Fe     55.84500  Fe.pbe-sp-van.UPF
>
> P      30.97376  P.pbe-n-van.UPF
>
> O      15.99940  O.pbe-rrkjus.UPF
>
>
>
> ATOMIC_POSITIONS {angstrom}
>
> Li      0.000000   0.000000   0.000000
>
> Li      2.375978   3.043675   5.225334
>
> Li      2.375978   0.000000   5.225334
>
> Li      0.000000   3.043675   0.000000
>
> Fe      2.263451   4.565513   8.168085
>
> Fe      4.639429   1.521838   7.507916
>
> Fe      0.112526   4.565513   2.942751
>
> Fe      2.488504   1.521838   2.282583
>
> P       2.760801   4.565513   0.989699
>
> P       0.384823   1.521838   4.235635
>
> P       4.367133   4.565513   6.215033
>
> P       1.991155   1.521838   9.460968
>
> O       1.223358   4.565513   1.010109
>
> O       1.359820   2.756426   8.719891
>
> O       1.359820   0.287250   8.719891
>
> O       3.528598   1.521838   9.440558
>
> O       1.396395   1.521838   0.446421
>
> O       3.599335   1.521838   4.215225
>
> O       3.392136   5.800101   1.730777
>
> O       3.735797   3.330925   6.956111
>
> O       1.016158   2.756426   3.494557
>
> O       1.016158   0.287250   3.494557
>
> O       3.355560   4.565513  10.004246
>
> O       3.772373   4.565513   4.778913
>
> O       0.979582   1.521838   5.671755
>
> O       1.152620   4.565513   6.235443
>
> O       3.735797   5.800101   6.956111
>
> O       3.392136   3.330925   1.730777
>
>
>
>
>
> And last few lines of output file are
>
> End of self-consistent calculation
>
>
>
>           k = 0.0000 0.0000 0.0000 (  3575 PWs)   bands (ev):
>
>
>
>    -79.0030 -79.0023 -79.0019 -78.9839 -45.2255 -45.2252 -45.2234 -45.2059
>
>    -45.1487 -45.1485 -45.1473 -45.1309 -44.9335 -44.9334 -44.9323 -44.9147
>
>    -15.3106 -15.1682 -15.0711 -14.9777 -12.6234 -12.6211 -12.5757 -12.5494
>
>    -12.5403 -12.5249 -12.5217 -12.4742 -12.4125 -12.3868 -12.3624 -12.3553
>
>     -2.5810  -2.1773  -2.1375  -2.0024  -0.9665  -0.6145  -0.5687  -0.4843
>
>     -0.1764  -0.1015   0.1339   0.2764   0.3243   0.3686   0.4349   0.4662
>
>      1.5656   1.9325   2.0392   2.0667   2.2272   2.2489   2.2587   2.2709
>
>      2.3867   2.5635   2.7745   2.9206   3.0322   3.0398   3.1022   3.1224
>
>      3.1659   3.1803   3.3798   3.4103   3.4174   3.5710   3.5829   3.5833
>
>      3.7764   3.8840   4.0086   4.0279   4.1655   4.3439   4.7040   4.8756
>
>      7.1039   7.1654   7.2249   7.2788   7.2957   7.3083   7.3196   7.3530
>
>      7.4510   7.4565   7.4624   7.4743   8.0415   8.0544   8.0944   8.3315
>
>      8.4876   8.5099   8.6253   8.6437  10.2682  12.0817  12.4471  13.1204
>
>     13.3931  13.9195  13.9632  13.9830  14.1605  14.3201  14.5312  14.7573
>
>     14.8969  14.9143  15.1052  15.6954  15.7497  16.0537  16.3593  16.5595
>
>     16.5733  16.6684  16.6782  16.7533  16.9513  17.0567  17.2068  17.6553
>
>     17.7728  17.7858  18.0596  18.5210
>
>
>
>      occupation numbers
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000   1.0000
>
>      0.9994   0.9993   0.9991   0.9987   0.0019   0.0013   0.0003   0.0000
>
>      0.0000   0.0000   0.0000   0.0000   0.0000   0.0000   0.0000   0.0000
>
>      0.0000   0.0000   0.0000   0.0000   0.0000   0.0000   0.0000   0.0000
>
>      0.0000   0.0000   0.0000   0.0000   0.0000   0.0000   0.0000   0.0000
>
>      0.0000   0.0000   0.0000   0.0000   0.0000   0.0000   0.0000   0.0000
>
>      0.0000   0.0000   0.0000   0.0000
>
>
>
>      the Fermi energy is     7.7636 ev
>
>
>
> !    total energy              =   -1586.80562557 Ry
>
>      Harris-Foulkes estimate   =   -1586.80581500 Ry
>
>      estimated scf accuracy    <       0.00024789 Ry
>
>
>
>      convergence has been achieved in  10 iterations
>
>
>
>
>
> Shoaib Muhammad, PhD
>
> Department of Energy Science
>
> Sungkyunkwan University
>
> South Korea



GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: <giuseppe.mattioli at ism.cnr.it>




More information about the users mailing list