[Pw_forum] GIPAW memory problem?

Aleksander Jaworski aleksander.jaworski at mmk.su.se
Mon May 12 17:18:32 CEST 2014


Dear QE users and developers,


I'm experiencing issues with running gipaw.x on the glass structures which
have been created from the classical MD simulations trajectories. 
On the glass pw.x is running smoothly and converging properly. Gipaw.x is
starting, but showing very strange behavior in terms of memory handling and
never terminating the job.
I'm testing very small clusters (35 atoms) so I don't think that hardware
resources are the limiting factors here.
Both pw.x and gipaw.x are runnig properly on the crystalline samples, so I
think I can exclude code compilation problems or pseudopotetnials bugs.
Problem is same when running on one core, so I don't think it is related
to parallelization. 
System is well converged with respect to e_cut and k_points.
I went across the mailing list archive and the manuals, and I cannot
conclude what is causing this problem.  

I would really appreciate if more experienced users could take a look on
the input and share with me hints and suggestions how to solve this
problem.


Below are the examples of the calculations done on 4 cores for quartz
(where pw.x and gipaw.x are working perfectly) and for 35 atoms glass
cluster (where pw.x is working properly and gipaw.x is never terminating
the job).

Gipaw.x output from the glass is there and both pw.x and gipaw.x inputs.


best regards,
Aleksander Jaworski

PhD student
Physical Chemistry Division
Stockholm University



pw.x on quartz:

VIRT  RES  SHR S  %CPU %MEM    TIME+  COMMAND           
227m  34m 8032 R  99.9  0.9   0:18.07 pw.x         
227m  34m 7324 R  99.9  0.9   0:18.08 pw.x              
227m  34m 7792 R  99.5  0.9   0:18.08 pw.x              
227m  35m 8536 R  92.5  0.9   0:17.39 pw.x

gipaw.x on quartz:

VIRT  RES  SHR S  %CPU %MEM    TIME+  COMMAND           
233m  39m 7560 R  99.9  1.0   0:22.84 gipaw.x           
233m  39m 7176 R  99.9  1.0   0:22.97 gipaw.x           
233m  39m 7072 R  99.5  1.0   0:22.98 gipaw.x           
233m  38m 6316 R  99.5  1.0   0:23.19 gipaw.x

pw.x on glass fragment 35 atoms:

VIRT  RES  SHR S  %CPU %MEM    TIME+  COMMAND           
386m 179m 6972 R  99.9  4.6  27:04.91 pw.x              
390m 189m 6164 R  99.9  4.8  27:05.81 pw.x              
390m 190m 6740 R  99.5  4.8  27:03.19 pw.x              
393m 194m 7428 R  99.2  4.9  26:50.74 pw.x

gipaw.x on glass fragment 35 atoms:

 VIRT  RES  SHR S  %CPU %MEM    TIME+  COMMAND           
 294m  23m 5976 R  99.9  0.6  88:09.55 gipaw.x           
12.3g  22m 6280 R  99.5  0.6  88:08.72 gipaw.x  <--- ???????        
 294m  23m 6080 R  99.5  0.6  88:09.16 gipaw.x           
 294m  22m 4936 R  99.5  0.6  88:06.15 gipaw.x 

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
gipaw.x output from my glass structure
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

Program GIPAW v.5.0.2 (svn rev. 9656) starts on  9May2014 at 21:41:13 

     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); 
          URL http://www.quantum-espresso.org", 
     in publications or presentations arising from this work. More details
at 
     http://www.quantum-espresso.org/quote.php 

     Parallel version (MPI), running on     4 processors 
     R & G space division:  proc/nbgrp/npool/nimage =       4 

   Info: using nr1, nr2, nr3 values from input 

   Info: using nr1s, nr2s, nr3s values from input 

     IMPORTANT: XC functional enforced from input : 
     Exchange-correlation      =  SLA  PW   PBX  PBC ( 1 4 3 4 0) 
     EXX-fraction              =        0.00 
     Any further DFT definition will be discarded 
     Please, verify this is what you really want 

               file Sc.pbe-spn-kjpaw_psl.0.2.3.UPF: wavefunction(s)  3P 3D
renormalized 
     WARNING: atomic wfc #  3 for atom type 2 has zero norm 
     WARNING: atomic wfc #  4 for atom type 2 has zero norm 
     WARNING: atomic wfc #  3 for atom type 3 has zero norm 
     WARNING: atomic wfc #  4 for atom type 3 has zero norm 

     Parallelization info 
     -------------------- 
     sticks:   dense  smooth     PW     G-vecs:    dense   smooth      PW 
     Min        2027     812    221               137553    34740    5083 
     Max        2028     813    222               137556    34751    5088 
     Sum        8109    3249    885               550217   138985   20341



........ and then nothing is going to happen, for days! Program is
printing this in the first seconds of the calculation and is not adding
anything until I will kill it, because of the lack of any progress.






Below are my inputs:
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
pw.x input
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
&control
    calculation = 'scf'
    restart_mode = 'from_scratch'
    prefix = 'Sc_glass'
    pseudo_dir = './pseudo/'
    outdir = './scratch/'
    tstress = .true.
    tprnfor = .true.
    verbosity = 'high'
/
&system
    ibrav = 0
    a = 7.5593007081
    nat = 35
    ntyp = 4
    ecutwfc = 50
    ecutrho = 500
    spline_ps = .true.
/
&electrons
    diagonalization = 'david'
    diago_thr_init = 1e-4
    mixing_mode = 'plain'
    mixing_beta = 0.7
    conv_thr =  1e-10
/
CELL_PARAMETERS  
 1.0000000000       0.0000000000        0.0000000000  
 0.0000000000       1.0000000000        0.0000000000  
 0.0000000000       0.0000000000        1.0000000000

ATOMIC_SPECIES
Sc     44.955     Sc.pbe-spn-kjpaw_psl.0.2.3.UPF
Al     26.982     Al.pbe-tm-gipaw-dc.UPF
Si     28.086     Si.pbe-tm-new-gipaw-dc.UPF
O      15.999     O.pbe-rrkjus-gipaw-dc.UPF

K_POINTS automatic
6 6 6 0 0 0

ATOMIC_POSITIONS angstrom 
Sc       3.653986436  -1.770430962   0.453388732
Sc       3.115400121   3.753142894  -2.621880013
Sc      -0.228578192  -1.909939933  -0.359045349
Al       1.515612487   0.598794798   1.286089223
Al      -2.142042574   0.099970299  -0.997825300
Al       1.457968354   3.343960195  -0.198592654
Al       0.142221475  -3.333943822   2.848542149
Al       2.430118347  -1.305929173   3.548774459
Si       2.140107709   0.526648186  -1.553243084
Si      -1.992462971  -1.301007123  -3.529125238
Si      -1.326216011   1.426125017   2.723247807
Si      -1.199730215   2.921918032  -2.063671047
Si       3.709225042   2.726202959   1.981252239
O       -1.902529337  -1.048857347   0.365493973
O       -3.664872460  -3.291948284   1.583056465
O        1.746530467  -0.105355750  -2.961930628
O        2.528356530  -0.694635328   1.893182099
O       -2.594445490   3.673588810  -1.938872319
O        1.128159783  -0.101570707  -0.346499313
O       -2.072151028   0.288350136   3.590230330
O       -1.332709251  -1.175862060  -2.030826061
O        3.366250358   2.407365161   3.508386106
O       -0.020611539  -3.717768926  -1.386394664
O        2.436173048  -2.829506418  -0.778460058
O        2.397436184   2.158266175   1.045311973
O       -1.202022794   1.486908837  -1.356904320
O       -0.781158682   2.726208527   3.761115686
O       -3.545735497  -1.823195853  -3.420669188
O        0.019501994   0.960865019   2.017730151
O       -2.410455796   1.966814312   1.616179899
O        0.422084553  -3.393089790   1.091238287
O        3.644536355   0.062847310  -1.003892882
O       -1.083072338  -2.149152584   3.048286539
O        2.016382220   2.165204491  -1.521542899
O        1.598040318  -2.865100753   3.709659988



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
gipaw.x input
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
&inputgipaw
    job = 'nmr'
    prefix = 'Sc_glass'
    tmp_dir = './scratch/'
    isolve = 0
    iverbosity = 1
    q_gipaw = 0.01
    spline_ps = .true.
    use_nmr_macroscopic_shape = .true.
/




More information about the users mailing list