Hi,
When I was running ph.x code to calculate the phonon dispersion, I ran it in parallel using the command " mpirun -np 48 ph.x -npool 48 < ph.in >ph.out". But in this case, there are more than one *.dvscf files generated (actually there are 48 of them). Therefore the pp.py does not work in this case since this script only deal with one *.dvscf file.
In the tutorial, it says "Note that all the calculations can also be performed in parallel using MPI, by issuing the following command: mpirun -np N ../../../../bin/pw.x -npool N < scf.in > scf.out". I am just using the same command but for phonon.
Could anyone tell me what should I do in this case?
Thanks in advance,
FC
running with npool for phonon calculation in ph.x
Moderator: stiwari
-
- Posts: 155
- Joined: Thu Jan 14, 2016 10:52 am
- Affiliation:
Re: running with npool for phonon calculation in ph.x
Hi,
You can run the phonon calculation with the parallelization you choose - it doesn't need to be the same as for the epw calculation. For example, you can choose np>npool to exploit G parallelization too.
There should only be one dvscf file for each q point that actually contains the info, and that's the one called dvscf1; the pp.py grabs that one.
Have you already tried using it? if yes, can you post the output of ls -lrt _pho/prefix.q_##/*dvscf* ? (## is any q point number)
Best
Carla
You can run the phonon calculation with the parallelization you choose - it doesn't need to be the same as for the epw calculation. For example, you can choose np>npool to exploit G parallelization too.
There should only be one dvscf file for each q point that actually contains the info, and that's the one called dvscf1; the pp.py grabs that one.
Have you already tried using it? if yes, can you post the output of ls -lrt _pho/prefix.q_##/*dvscf* ? (## is any q point number)
Best
Carla
Re: running with npool for phonon calculation in ph.x
carla.verdi wrote:Hi,
You can run the phonon calculation with the parallelization you choose - it doesn't need to be the same as for the epw calculation. For example, you can choose np>npool to exploit G parallelization too.
There should only be one dvscf file for each q point that actually contains the info, and that's the one called dvscf1; the pp.py grabs that one.
Have you already tried using it? if yes, can you post the output of ls -lrt _pho/prefix.q_##/*dvscf* ? (## is any q point number)
Best
Carla
Thank you for your reply.
Below is the out when using command "ls -lrt _pho/prefix.q_##/*dvscf*" In this case I used 96 cores. And there is 96 dvscf files.
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf78
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf16
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf92
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf88
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf80
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf51
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf20
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf8
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf42
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf24
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf17
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf94
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf76
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf30
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf12
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf10
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf96
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf71
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf7
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf69
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf59
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf48
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf19
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf90
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf86
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf84
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf66
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf52
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf27
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf18
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf15
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf14
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf91
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf54
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf53
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf50
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf40
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf4
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf32
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf28
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf26
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf21
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf2
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf93
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf9
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf89
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf82
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf77
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf65
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf63
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf57
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf55
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf5
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf3
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf22
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf95
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf85
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf83
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf74
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf62
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf61
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf60
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf6
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf58
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf47
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf46
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf38
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf23
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf13
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf11
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf87
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf81
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf79
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf75
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf73
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf70
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf68
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf64
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf56
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf49
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf44
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf41
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf37
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf36
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf35
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf72
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf67
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf45
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf43
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf39
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf34
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf33
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf31
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf29
-rw-r--r-- 1 fanchem cuuser 0 Feb 18 00:53 _ph0/a.q_2/a.dvscf25
-rw-r--r-- 1 fanchem cuuser 4608000 Feb 18 00:54 _ph0/a.q_2/a.dvscf1
Thank you,
Re: running with npool for phonon calculation in ph.x
Hi,
You only need the following file
rw-r--r-- 1 fanchem cuuser 4608000 Feb 18 00:54 _ph0/a.q_2/a.dvscf1
All the other files are empty.
Best,
Roxana
You only need the following file
rw-r--r-- 1 fanchem cuuser 4608000 Feb 18 00:54 _ph0/a.q_2/a.dvscf1
All the other files are empty.
Best,
Roxana
Roxana Margine
Associate Professor
Department of Physics, Applied Physics and Astronomy
Binghamton University, State University of New York
Associate Professor
Department of Physics, Applied Physics and Astronomy
Binghamton University, State University of New York
Re: running with npool for phonon calculation in ph.x
roxana wrote:Hi,
You only need the following file
rw-r--r-- 1 fanchem cuuser 4608000 Feb 18 00:54 _ph0/a.q_2/a.dvscf1
All the other files are empty.
Best,
Roxana
I just realized that. Thank you so much for your reply.
Best,
fanchem