[QE-users] 回复: Re: Error happens after setting max_seconds in turbo_davidson.x

508682179 at qq.com
Mon Apr 19 05:48:42 CEST 2021


Dear Iurri,


With ecut = 40, the restart files can be written successfully for both QE6.5 and QE6.7. But I afraid I have to set ecut as 80 as this is the value leads to the convergece of "band edge". If do the turbo davidson calculation with ecut = 40,  the result will be out of accuracy as this calculation is based on both occupied and unoccupied states, won't it?  


The problem is that the restart file cannot be written once ecut=80 used in scf calculation (even using QE6.7). Do you have any suggestion to solve this restart file problem without affecting the accuracy of the result. Thanks.






Best,
Weijie Zhou




----------------------------------------------------------------
Weijie Zhou
PhD student
University of Leeds
UK







------------------ 原始邮件 ------------------
发件人:                                                                                                                        "Iurii TIMROV"                                                                                    <users at lists.quantum-espresso.org>;
发送时间: 2021年4月17日(星期六) 晚上10:21
收件人: "users"<users at lists.quantum-espresso.org>;

主题: Re: [QE-users]Re: Error happens after setting max_seconds in	turbo_davidson.x



  
Dear Weijie,
 

 
 
I have run a test using QE6.7 and your inputs with reduced parameters on my workstations using 8 cores. The code works fine, the restart files are written successfully when the turbo_davidson.x code reached max_seconds. Below are my inputs, you can try them  as well.
 

 
 
Please check that you have enough disc space.
 

 
 
Greetings,
 
Iurii
 

 
 
  &control
 
     calculation='scf'
 
     restart_mode='from_scratch',
 
     prefix='MoO'
 
     pseudo_dir = '/scratch/timrov/q-e-qe-6.7MaX-Release/pseudo/'
 
     outdir='/scratch/timrov/q-e-qe-6.7MaX-Release/tempdir/'
 
  /
 
  &system
 
     a                         =   2.00000e+01
 
     ecutwfc                   =   30
 
     ibrav                     = 1
 
     nat                       = 34
 
     ntyp                      = 2
 
     nbnd                      =250
 
     tot_charge= -4
 
     assume_isolated='martyna-tuckerman'
 
  /
 
  &electrons
 
     conv_thr =  1.0d-8
 
  /
 
 ATOMIC_SPECIES
 
 Mo  95.94    Mo.upf
 
 O  15.999    O.upf
 
 ATOMIC_POSITIONS {angstrom}
 
 Mo      0.011800  -0.882450  -2.732125
 
 Mo      0.210675  -2.315000   1.646750
 
 Mo      1.961325  -2.892175  -1.042350
 
 Mo     -1.743275  -0.383075  -0.047150
 
 O       0.170525  -1.302200  -0.431375
 
 O       1.833025  -2.939550   0.887000
 
 O       1.668550  -1.814000  -2.603400
 
 O       0.852375  -4.156450  -1.364525
 
 O       3.494400  -3.581575  -1.331675
 
 O      -1.164775  -0.859925   1.754125
 
 O      -1.326825   0.320475  -1.821825
 
 O      -2.853125   0.892925   0.413600
 
 O      -2.723075  -1.719050  -0.429725
 
 O      -0.880150  -3.578800   1.175650
 
 O       0.468325  -2.590150   3.305325
 
 O       0.158250  -0.119825  -4.231600
 
 O      -1.075000  -2.186500  -3.000775
 
 Mo     -0.011825   0.882450   2.732100
 
 Mo     -0.210700   2.315000  -1.646775
 
 Mo     -1.961350   2.892175   1.042325
 
 Mo      1.743275   0.383075   0.047100
 
 O      -0.170550   1.302225   0.431325
 
 O      -1.833050   2.939575  -0.887025
 
 O      -1.668550   1.814000   2.603375
 
 O      -0.852375   4.156450   1.364475
 
 O      -3.494400   3.581575   1.331650
 
 O       1.164750   0.859950  -1.754175
 
 O       1.326800  -0.320450   1.821775
 
 O       2.853125  -0.892925  -0.413625
 
 O       2.723075   1.719050   0.429700
 
 O       0.880125   3.578800  -1.175675
 
 O      -0.468325   2.590150  -3.305350
 
 O      -0.158275   0.119850   4.231575
 
 O       1.075000   2.186500   3.000725
 
 K_POINTS {gamma}
 
 
 
 
 
 
 
  &lr_input
 
    prefix = 'MoO',
 
    outdir = '/scratch/timrov/q-e-qe-6.7MaX-Release/tempdir/',
 
    restart = .false.
 
    max_seconds = 600
 
  /
 
  &lr_dav
 
   if_dft_spectrum= .false.
 
   num_eign = 6,
 
   num_init = 12,
 
   num_basis_max = 40,
 
   residue_conv_thr = 1.0D-4,
 
   start  = 0.0,
 
   finish = 6,
 
   step   = 2.0D-4,
 
   broadening = 0.005,
 
   reference = 0.5,
 
   p_nbnd_occ  = 8,
 
   p_nbnd_virt = 8,
 
   poor_of_ram  = .false.
 
   poor_of_ram2 = .true.
 
 
 
 /
 

 
     --
 Dr. Iurii TIMROV
 Senior Research Scientist
  Theory and Simulation of Materials (THEOS)
  Swiss Federal Institute of Technology Lausanne (EPFL)
 
   CH-1015 Lausanne, Switzerland
 +41 21 69 34 881
   http://people.epfl.ch/265334
 
 
 
 
 
 
 From: users <users-bounces at lists.quantum-espresso.org> on behalf of 飘 <508682179 at qq.com>
 Sent: Friday, April 16, 2021 8:03:37 PM
 To: users
 Subject: [QE-users] Re: Error happens after setting max_seconds in turbo_davidson.x  
 
  Dear Iurii,
 
 
 I used example 8 of TDDFPT for testing restart work under QE6.5 and it works fine as well. So I am afraid it is not the version problem.
 
 
 Best, 
   Weijie Zhou
 
 
 
 
 
 
  ----------------------------------------------------------------
 Weijie Zhou
 PhD student
 University of Leeds
 UK
 
 
 
 
 ------------------ 原始邮件 ------------------
  发件人: "Iurii TIMROV" <users at lists.quantum-espresso.org>;
 发送时间: 2021年4月16日(星期五) 下午5:34
 收件人: "users"<users at lists.quantum-espresso.org>;
 
 主题: Re: [QE-users]RE: Error happens after setting max_seconds in turbo_davidson.x
 
 
 
  
Dear Weijie,
 
 I made a quick test using example 8 of TDDFPT and the restart works fine. Therefore, try to use QE6.7 and try example 8. If it works for you, then try it for your system (for the sake of testing reduce the cutoff, num_eign, num_init, max_seconds). 
 

 
 
Greetings,
 
Iurii
 
 

 
     --
 Dr. Iurii TIMROV
 Senior Research Scientist
  Theory and Simulation of Materials (THEOS)
  Swiss Federal Institute of Technology Lausanne (EPFL)
 
   CH-1015 Lausanne, Switzerland
 +41 21 69 34 881
   http://people.epfl.ch/265334
 
 
 
 
 
 
 From: users <users-bounces at lists.quantum-espresso.org> on behalf of 飘 <508682179 at qq.com>
 Sent: Monday, April 12, 2021 10:49:51 AM
 To: users
 Subject: [QE-users] RE: Error happens after setting max_seconds in turbo_davidson.x  
 
   Dear Iurii,
 
 
 Thanks for your reply and reminder. I am using QE 6.5. And you can see the input & output files in:
 
 
 https://drive.google.com/drive/folders/1djSkUZ2MkF0nKNntQgZJ8wt3b3IpmHs-?usp=sharing
  
 
 I am sure there is  enough space and time for writing data for restart files after double check.  
 
 
 
 
 Best,
 
 
 Weijie Zhou
 
 
 
 
 
 
  ----------------------------------------------------------------
 Weijie Zhou
 PhD student
 University of Leeds
 UK
 
 
 
  
 
 
 
 ------------------ 原始邮件 ------------------
  发件人: "Iurii TIMROV" <users at lists.quantum-espresso.org>;
 发送时间: 2021年4月11日(星期天) 下午5:23
 收件人: "users"<users at lists.quantum-espresso.org>;
 
 主题: Re: [QE-users] Error happens after setting max_seconds in turbo_davidson.x
 
 
 
  
Please read carefully how to ask questions on the pw_forum:  https://www.quantum-espresso.org/forum
 
 
In particular:
 
- which version of QE do you use?
 
- provide all input and output files (so that we can reproduce your problem)
 

 
 
Moreover: 
 
- make sure you have enough disc space to write the data for restart
 
- make sure that the code has enough time to write the data to disc (i.e. compare max_seconds with the total time of the job)
 

 
 
Once all this is clarified, the QE developers can have a closer look at your problem.
 

 
 
HTH
 

 
 
Iurii
 

 
     --
 Dr. Iurii TIMROV
 Senior Research Scientist
  Theory and Simulation of Materials (THEOS)
  Swiss Federal Institute of Technology Lausanne (EPFL)
 
   CH-1015 Lausanne, Switzerland
 +41 21 69 34 881
   http://people.epfl.ch/265334
 
 
 
 
 
 
 From: users <users-bounces at lists.quantum-espresso.org> on behalf of 飘 <508682179 at qq.com>
 Sent: Sunday, April 11, 2021 3:47:21 AM
 To: users
 Subject: [QE-users] Error happens after setting max_seconds in turbo_davidson.x  
 
  Dear QE users,
 
 
 I am trying to use "restart" function for turbo Davidson calculation. So the "max_second" was specified in the input files of turbo_davidson.x which is shown as below:
 
 
  &lr_input
     prefix='Mo_O',
     outdir='../../../tmp_Mo8O26_davidson_size20_e-4_nosmearing_T2'
     restart=.false.
     max_seconds = 144000,
 /
 &lr_dav
 if_dft_spectrum= .false.
 num_eign = 16,
   num_init = 32,
   num_basis_max = 80,
   residue_conv_thr = 1.0D-4,
   start  = 0.0,
   finish = 6,
   step   = 2.0D-4,
   broadening = 0.005,
   reference = 0.5,
   p_nbnd_occ  = 8,
   p_nbnd_virt = 8,
   poor_of_ram  = .false.
    poor_of_ram2 = .true.
  
 
 But error happened at the end of calculation (max_second is reached) when writing data for restart. The error message: 
   %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine diropn (22):
     error opening ../../../tmp_Mo8O26_davidson_size20_e-4_nosmearing_T2/Mo_O.restart_davidson_vec_b.3
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
      stopping ...
  
 
 It will be very appreciated if you can solve my problem for me. Thanks.
 
 
 Best,
 Weijie Zhou
 
 
 
 
 
 
 ----------------------------------------------------------------
 Weijie Zhou
 PhD student
 University of Leeds
 UK
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20210419/43fc845a/attachment.html>


More information about the users mailing list