Page 1 of 1

epw.x just does nothing after "G-vector sticks info"

Posted: Thu Jan 18, 2018 9:44 am
by tessw
Hi,

I am trying now for several days to get epw up and running. The bulk fcc PB example did work properly, so epw should be working. I then adjusted the input files accordingly to suit my needs. Sadly, epw now gets stuck after the line "G-vector sticks info" and does nothing anymore, not even output some kind of error message. Has anyone experienced the same problem or could maybe spot some error in one of my input files (see below)?

I tried running with 24 cores and mpirun, but also with only one core. If mpirun was used, "mpirun -np 24 pw.x -npool 24 < scf.in" and so one was the way I tried, according to the examples online. All the other problems I had could by solved by adjusting the input files with the help of this forum or the mailing list, but this time I am stuck, as the code does not break and there is no error message but it is just stuck.

I highly appreciate any help! Thanks :)
Tobias

Code: Select all

scf.in
&control
    calculation   ='scf',
    prefix        ='kto',
    pseudo_dir    = '../../pp/',
    outdir        ='./',
    tprnfor       = .true.,
    tstress       = .true.,   
    etot_conv_thr = 1.0d-4
    forc_conv_thr = 1.0d-4
    wf_collect    = .true.
 /
 &system   
    ibrav       = 1 ,
    A           = 4.0109        ! (Angstrom)
    nat         =  5 ,          ! total number of atoms in UC
    ntyp        = 3 ,           ! number of unique atoms in UC
    ecutwfc     = 60 ,          ! (Ry)
    ecutrho     = 720 ,         ! (Ry)
    occupations = 'smearing',   ! gaussian smearing for metals
    degauss     = 0.05          ! (Ry) gaussian spreading in metals
    noncolin    = .false.       ! std: .false.
    lspinorb    = .false.       ! std: .false.
 /
 &electrons
    conv_thr    = 1.0d-8        ! std: 1d-6
    mixing_ndim = 8             ! std: 8
    mixing_beta = 0.7           ! std: 0.7
    diagonalization = 'cg'
 /
ATOMIC_SPECIES
    K     39.0983  K.pbe-spn-rrkjus_psl.1.0.0.UPF
    Ta   180.9479  Ta.pbe-spn-rrkjus_psl.0.2.UPF
    O     15.9990  O.pbe-n-rrkjus_psl.0.1.UPF
ATOMIC_POSITIONS crystal
    K   0.000000000000000   0.000000000000000   0.000000000000000
    Ta  0.500000000000000   0.500000000000000   0.500000000000000
    O   0.000000000000000   0.500000000000000   0.500000000000000
    O   0.500000000000000   0.500000000000000   0.000000000000000
    O   0.500000000000000   0.000000000000000   0.500000000000000
K_POINTS AUTOMATIC
8 8 8 1 1 1

Code: Select all

ph.in
&inputph
    recover  = .false.
    prefix   = 'kto',
    outdir   = './'
    fildyn   = 'kto.dyn',
    fildvscf = 'dvscf'
    ldisp    = .true.,
    nq1      = 4,
    nq2      = 4,
    nq3      = 4,
    tr2_ph   = 1.0d-14,  ! std: 1.0d-12
 /

Code: Select all

scf.in
 &control
    calculation   ='scf',
    restart_mode = 'from_scratch' ,
    prefix        ='kto',
    pseudo_dir    = '../../pp/',
    outdir        ='./',
    tprnfor       = .true.,
    tstress       = .true.,   
    etot_conv_thr = 1.0d-4
    forc_conv_thr = 1.0d-4
    wf_collect    = .true.
 /
 &system   
    ibrav       = 1 ,
    A           = 4.0109        ! (Angstrom)
    nat         =  5 ,          ! total number of atoms in UC
    ntyp        = 3 ,           ! number of unique atoms in UC
    ecutwfc     = 60 ,          ! (Ry)
    ecutrho     = 720 ,         ! (Ry)
    occupations = 'smearing',   ! gaussian smearing for metals
    degauss     = 0.05          ! (Ry) gaussian spreading in metals
    noncolin    = .false.       ! std: .false.
    lspinorb    = .false.       ! std: .false.
 /
 &electrons
    conv_thr    = 1.0d-8        ! std: 1d-6
    mixing_ndim = 8             ! std: 8
    mixing_beta = 0.7           ! std: 0.7
    diagonalization = 'cg'
 /
ATOMIC_SPECIES
    K     39.0983  K.pbe-spn-rrkjus_psl.1.0.0.UPF
    Ta   180.9479  Ta.pbe-spn-rrkjus_psl.0.2.UPF
    O     15.9990  O.pbe-n-rrkjus_psl.0.1.UPF
ATOMIC_POSITIONS crystal
    K   0.000000000000000   0.000000000000000   0.000000000000000
    Ta  0.500000000000000   0.500000000000000   0.500000000000000
    O   0.000000000000000   0.500000000000000   0.500000000000000
    O   0.500000000000000   0.500000000000000   0.000000000000000
    O   0.500000000000000   0.000000000000000   0.500000000000000
K_POINTS AUTOMATIC
8 8 8 1 1 1

Code: Select all

nscf.in
 &control
    calculation ='nscf',
    prefix      ='kto',
    pseudo_dir  = '../../pp/',
    outdir      ='./',
    tprnfor     = .true.,
    tstress     = .true.,
 /
 &system   
    ibrav       = 1 ,
    A           = 4.0109       ! (Angstrom)
    nat         =  5 ,         ! total number of atoms in UC
    ntyp        = 3 ,          ! number of unique atoms in UC
    ecutwfc     = 60 ,         ! (Ry)
    ecutrho     = 720 ,        ! (Ry)
    occupations = 'smearing',  ! gaussian smearing for metals
    degauss     = 0.05         ! (Ry) gaussian spreading in metals
    noncolin    = .false.      ! std: .false.
    lspinorb    = .false.      ! std: .false.
    nbnd        = 35 ,
    nosym       = .true.
    noinv       = .true.
 /
 &electrons
    conv_thr    = 1.0d-8        ! std: 1d-6
    mixing_ndim = 8             ! std: 8
    mixing_beta = 0.7           ! std: 0.7
    diagonalization = 'cg'
 /
ATOMIC_SPECIES
    K     39.0983  K.pbe-spn-rrkjus_psl.1.0.0.UPF
    Ta   180.9479  Ta.pbe-spn-rrkjus_psl.0.2.UPF
    O     15.9990  O.pbe-n-rrkjus_psl.0.1.UPF
ATOMIC_POSITIONS crystal
    K   0.000000000000000   0.000000000000000   0.000000000000000
    Ta  0.500000000000000   0.500000000000000   0.500000000000000
    O   0.000000000000000   0.500000000000000   0.500000000000000
    O   0.500000000000000   0.500000000000000   0.000000000000000
    O   0.500000000000000   0.000000000000000   0.500000000000000
K_POINTS crystal
64
  0.00000000 0.00000000 0.00000000 0.01562500
  0.00000000 0.00000000 0.25000000 0.01562500
  0.00000000 0.00000000 0.50000000 0.01562500
  0.00000000 0.00000000 0.75000000 0.01562500
  0.00000000 0.25000000 0.00000000 0.01562500
  0.00000000 0.25000000 0.25000000 0.01562500
......
  0.75000000 0.50000000 0.75000000 0.01562500
  0.75000000 0.75000000 0.00000000 0.01562500
  0.75000000 0.75000000 0.25000000 0.01562500
  0.75000000 0.75000000 0.50000000 0.01562500
  0.75000000 0.75000000 0.75000000 0.01562500

Code: Select all

epw.in
--
&inputepw
  prefix      = 'kto',
  outdir      = './'

  elph        = .true.
  kmaps       = .false.
  epbwrite    = .true.
  epbread     = .false.

  epwwrite    = .true.
  epwread     = .false.

  nbndsub     = 30
  nbndskip    = 5
  fsthick     = 30

  wannierize  = .true.
  dis_win_max = 15
  dis_win_min = -5
  dis_froz_max= 5
  dis_froz_min= -5
  proj(1)     = 'Ta:l=2'
  proj(2)     = 'O:l=1'

  wdata(1) = 'bands_plot = .true.'
  wdata(2) = 'begin kpoint_path'
  wdata(3) = 'G 0.00 0.00 0.00 X 0.50 0.00 0.00'
  wdata(4) = 'X 0.50 0.00 0.00 M 0.50 0.50 0.00'
  wdata(5) = 'M 0.50 0.50 0.00 G 0.00 0.00 0.00'
  wdata(6) = 'G 0.00 0.00 0.00 R 0.50 0.50 0.50'
  wdata(7) = 'R 0.50 0.50 0.50 X 0.50 0.00 0.00'
  wdata(8) = 'end kpoint_path'
  wdata(9) = 'bands_plot_format = gnuplot'

  iverbosity  = 0

  elecselfen  = .true.
  phonselfen  = .true.

  parallel_k  = .true.
  parallel_q  = .false.

  eptemp      = 0.005 ! K
  degaussw    = 0.025 ! eV
  degaussq    = 0.050 !meV

  a2f         = .true.

  dvscf_dir   = '../phonons/save'

  nkf1         = 10
  nkf2         = 10
  nkf3         = 10

  nqf1         = 10
  nqf2         = 10
  nqf3         = 10

  nk1         = 4
  nk2         = 4
  nk3         = 4

  nq1         = 4
  nq2         = 4
  nq3         = 4

 /
  10 cartesian
       0.000000000   0.000000000   0.000000000
       0.000000000   0.000000000   0.250000000
       0.000000000   0.000000000  -0.500000000
       0.000000000   0.250000000   0.250000000
       0.000000000   0.250000000  -0.500000000
       0.000000000  -0.500000000  -0.500000000
       0.250000000   0.250000000   0.250000000
       0.250000000   0.250000000  -0.500000000
       0.250000000  -0.500000000  -0.500000000
      -0.500000000  -0.500000000  -0.500000000


Code: Select all

epw.out                                                                                      
                                       ``:oss/                                       
                           `.+s+.     .+ys--yh+     `./ss+.                           
                          -sh//yy+`   +yy   +yy    -+h+-oyy                           
                          -yh- .oyy/.-sh.   .syo-.:sy-  /yh                           
                 `.-.`    `yh+   -oyyyo.     `/syys:    oys      `.`                 
               `/+ssys+-` `sh+      `                   oys`   .:osyo`               
               -yh- ./syyooyo`                          .sys+/oyo--yh/               
               `yy+    .-:-.                             `-/+/:`  -sh-               
                /yh.                                              oys                 
          ``..---hho---------`   .---------..`      `.-----.`    -hd+---.             
       `./osmNMMMMMMMMMMMMMMMs. +NNMMMMMMMMNNmh+.   yNMMMMMNm-  oNMMMMMNmo++:`       
       +sy--/sdMMMhyyyyyyyNMMh- .oyNMMmyyyyyhNMMm+` -yMMMdyyo:` .oyyNMMNhs+syy`       
       -yy/   /MMM+.`-+/``mMMy-   `mMMh:`````.dMMN:` `MMMy-`-dhhy```mMMy:``+hs       
        -yy+` /MMMo:-mMM+`-oo/.    mMMh:     `dMMN/`  dMMm:`dMMMMy..MMMo-.+yo`       
         .sys`/MMMMNNMMMs-         mMMmyooooymMMNo:   oMMM/sMMMMMM++MMN//oh:         
          `sh+/MMMhyyMMMs- `-`     mMMMMMMMMMNmy+-`   -MMMhMMMsmMMmdMMd/yy+           
    `-/+++oyy-/MMM+.`/hh/.`mNm:`   mMMd+/////:-.`      NMMMMMd/:NMMMMMy:/yyo/:.`     
   +os+//:-..-oMMMo:--:::-/MMMo. .-mMMd+---`           hMMMMN+. oMMMMMo. `-+osyso:`   
   syo     `mNMMMMMNNNNNNNNMMMo.oNNMMMMMNNNN:`         +MMMMs:`  dMMMN/`     ``:syo   
   /yh`     :syyyyyyyyyyyyyyyy+.`+syyyyyyyyo:`         .oyys:`   .oyys:`        +yh   
   -yh-        ````````````````    `````````              ``        ``          oys   
   -+h/------------------------::::::::://////++++++++++++++++++++++///////::::/yd:   
   shdddddddddddddddddddddddddddddhhhhhhhhyyyyyssssssssssssssssyyyyyyyhhhhhhhddddh`   
                                                                                     
  S. Ponce, E. R. Margine, C. Verdi, and F. Giustino,                                 
                                                Comput. Phys. Commun. 209, 116 (2016)
                                                                                     

     Program EPW v.4.3.0 (svn rev. 14038) starts on 18Jan2018 at 10:12: 5

     This program is part of the open-source Quantum ESPRESSO suite
     for quantum simulation of materials; please cite
         "P. Giannozzi et al., J. Phys.:Condens. Matter 21 395502 (2009);
         "P. Giannozzi et al., J. Phys.:Condens. Matter 29 465901 (2017);
          URL http://www.quantum-espresso.org",
     in publications or presentations arising from this work. More details at
     http://www.quantum-espresso.org/quote

     Parallel version (MPI), running on     1 processors

     MPI processes distributed on     1 nodes

     Reading data from directory:
     ./kto.save/

     IMPORTANT: XC functional enforced from input :
     Exchange-correlation      = PBE ( 1  4  3  4 0 0)
     Any further DFT definition will be discarded
     Please, verify this is what you really want

               file K.pbe-spn-rrkjus_psl.1.0.0.UPF: wavefunction(s)  3S renormalized
 
     G-vector sticks info
     --------------------
     sticks:   dense  smooth     PW     G-vecs:    dense   smooth      PW
     Sum        3297    1101    357               142009    27369    5041
 

     Check: negative/imaginary core charge=   -0.000007    0.000000

Re: epw.x just does nothing after "G-vector sticks info"

Posted: Sat Jan 20, 2018 2:16 pm
by sponce
Dear tessw,

I'm afraid only norm-conserving psp are supported in EPW at the moment (this is something we would like to support in the realtively near future).

Best,
Samuel

Re: epw.x just does nothing after "G-vector sticks info"

Posted: Mon Jan 29, 2018 2:47 pm
by tessw
Dear Samuel,

thank you very much for your fast reply! I was not aware of that and now changed to norm conserving pseudo potentials, but sadly the above mentioned error persists. Do you have any idea why this could occur?

Best,
Tobias

Re: epw.x just does nothing after "G-vector sticks info"

Posted: Mon Jan 29, 2018 6:15 pm
by sponce
Dear Tobias,

Can you try one EPW test to make sure everything is ok. You can find it in QE/test-suite/epw_base/ for example.

If this does not work in parallel, then you have an issue with the compilation.
If it does work, then a possible issue might be memory. What is the size of your wfc/dvscf etc ?

Best,
Samuel

Re: epw.x just does nothing after "G-vector sticks info"

Posted: Mon Jan 29, 2018 6:16 pm
by Vahid
I once had a similar issue. One reason was that for norm-conserving PSP, ecutrho=4*ecutwfc. I chose ecutrho larger than 4*ecutwfc and EPW got stuck. Did you change ecutrho after you switched to norm-conserving PSP?

Cheers,

Vahid

Vahid Askarpour
Department of Physics and Atmospheric Science
Dalhousie University,
Halifax, NS, Canada

Re: epw.x just does nothing after "G-vector sticks info"

Posted: Mon Jan 29, 2018 6:21 pm
by sponce
Well spotted Vahid. Its probably the issue indeed.

Just remove ecutrho from scf and nscf.in

Best,
Samuel

Re: epw.x just does nothing after "G-vector sticks info"

Posted: Tue Jan 30, 2018 11:32 am
by tessw
Dear Samuel and Vahid,

thank you very much for your inputs!

To answer your questions, the wfc files from nscf are around 4 MB and the dvscf from ph around 100 MB each. As I tried with 96 GB RAM on 12 cores and 30 GB RAM on one core, I think that was not the problem. In the beginning I also first did the whole pb-example from the homepage, which did work properly, before I changed the inputs to suit my needs.

But apparently there was a combined issue with compilation, parallelization and/or the ecutrho. I finally have working environments on two machines.
On one machine I recompiled QE with intel/13.3.163 and openmpi/1.6.5 and changed the c-flag from -o3 to -o2 and now it does work on one core, but not on 12 (even with the high ecutrho).
On another machine it now even works on 12 cores, compiled as well with -o2 instead of -o3, but with open_mpi/2.1.2 and intel/2018.0 and the scf and nscf calculations done with the normal ecutrho=4*ecutwfc.

If I can find the actual cause why it did not work in the first place, I will post it here.

Thanks again for your quick and helpful replies! :)