[QE-users] Cholesky error in NSCF calculation

[Kayahan Saritas]

Dear QE Users and developers, 

I am trying to run a relativistic NSCF calculation of a slab using the 
non-relativistic collinear charge density. I see that a similar problem has 
been posted about 4 years ago, but no solution was provided.

https://lists.quantum-espresso.org/pipermail/users/2018-July/040944.html 


I have two NSCF calculations where spin quantization axes are in x and z 
directions to calculate the magnetic anisotropy energy. The initial SCF 
calculation, spin quantization axis in z calculation converges with no issues. 
Only the quantization axis in x calculation fails. As previously reported in 
the link, I have tried multiple optimizers (cg, paro, rmm-davidson, david), but 
none of them worked. Since that post was about 4 years old, I was wondering if 
there have been any solution to this since then. I have my input provided below 
in case you would like to have a look. 

Thanks,

Kayahan Saritas
Yale University


   calculation = 'nscf'
   disk_io = 'low'
   outdir  = 'pwscf_output'
   prefix  = 'pwscf'
   pseudo_dir  = './'
   restart_mode= 'from_scratch'
   tprnfor = .true.
   tstress = .true.
   verbosity   = 'high'
   wf_collect  = .true.
/


   angle1(1)   = 90
   angle1(2)   = 90
   angle1(3)   = 90
   angle1(4)   = 90
   angle2(1)   = 0
   angle2(2)   = 0
   angle2(3)   = 0
   angle2(4)   = 0
   degauss = 0.001
   ecutrho = 720
   ecutwfc = 60
   ibrav   = 0
   lforcet = .true.
   lspinorb= .true.
   nat = 53
   nbnd= 960
   noinv   = .true.
   noncolin= .true.
   nosym   = .true.
   nosym_evc   = .true.
   ntyp= 4
   occupations = 'smearing'
   smearing= 'm-v'
   starting_magnetization(1) = 1
   tot_charge  = 0
/


   conv_thr= 1e-08
   diagonalization = 'paro'
   electron_maxstep = 1000
   mixing_beta = 0.2
   mixing_mode = 'plain'
/


ATOMIC_SPECIES
   Fe 55.8450006 Fe.rel-pbe-spn-rrkjus_psl.1.0.0.UPF
   O  15.999 O.pbe-n-rrkjus_psl.1.0.0.UPF
   Pd 106.42 Pd.rel-pbe-spn-rrkjus_psl.1.0.0.UPF
   Si 28.086 Si.rel-pbe-n-rrkjus_psl.1.0.0.UPF

ATOMIC_POSITIONS bohr
…

K_POINTS automatic
   10 10 1  0 0 0

CELL_PARAMETERS bohr
10.44450878   0.   0.
-5.5439   9.04521004   0.
 0.   0.  76.75114374





smime.p7s
Description: S/MIME cryptographic signature
___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] cholesky Error

[Laurens Siemons]

If your sure there's nothing wrong with your structure or your input 
parameters, then you can try to  add 'nd 1' to your script when you submit it 
like:

pw.x -nd 1 -inp name.rx.in > name.rx.out

This worked for me.
 Laurens Siemons

Van: users <users-boun...@lists.quantum-espresso.org> namens Amin Mirzai 
<amin.mir...@mek.lth.se>
Verzonden: maandag 9 april 2018 12:49
Aan: Quantum Espresso users Forum
Onderwerp: Re: [QE-users] cholesky Error


Thanks for the hint, I understand this topic has already been discussed. 
However, after doing all possible variations I still get the same error.

Anyhow, I believe there is not any explicit answer to this issue.


Regards,

Amin



From: users <users-boun...@lists.quantum-espresso.org> on behalf of Lorenzo 
Paulatto <paul...@gmail.com>
Sent: Monday, April 9, 2018 10:00:13 AM
To: users@lists.quantum-espresso.org
Subject: Re: [QE-users] cholesky Error

On 09/04/18 09:57, Amin Mirzai wrote:
>   %%
>   Error in routine  cdiaghg (251):
>problems computing cholesky
>   %
>
This is one of the most regularly asked questions. This error typically
comes from bad geometry (i.e. ions core regions overlapping), bad
pseudopotentials (should not happen with anything from usual libraries)
or problematic input parameters with advanced features (i.e. lda+U, exx,...)

kind regards


--
Lorenzo Paulatto - Paris
___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] cholesky Error

[Amin Mirzai]

Thanks for the hint, I understand this topic has already been discussed. 
However, after doing all possible variations I still get the same error.

Anyhow, I believe there is not any explicit answer to this issue.


Regards,

Amin



From: users <users-boun...@lists.quantum-espresso.org> on behalf of Lorenzo 
Paulatto <paul...@gmail.com>
Sent: Monday, April 9, 2018 10:00:13 AM
To: users@lists.quantum-espresso.org
Subject: Re: [QE-users] cholesky Error

On 09/04/18 09:57, Amin Mirzai wrote:
>   %%
>   Error in routine  cdiaghg (251):
>problems computing cholesky
>   %
>
This is one of the most regularly asked questions. This error typically
comes from bad geometry (i.e. ions core regions overlapping), bad
pseudopotentials (should not happen with anything from usual libraries)
or problematic input parameters with advanced features (i.e. lda+U, exx,...)

kind regards


--
Lorenzo Paulatto - Paris
___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users
___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] cholesky Error

[Lorenzo Paulatto]

On 09/04/18 09:57, Amin Mirzai wrote:

  %%
  Error in routine  cdiaghg (251):
   problems computing cholesky
  %

This is one of the most regularly asked questions. This error typically 
comes from bad geometry (i.e. ions core regions overlapping), bad 
pseudopotentials (should not happen with anything from usual libraries) 
or problematic input parameters with advanced features (i.e. lda+U, exx,...)


kind regards


--
Lorenzo Paulatto - Paris
___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

[QE-users] cholesky Error

[Amin Mirzai]

Dear Developers and Users,


I have been trying to vc-relax a crystal, but every time I have encountered 
with the following error:

 %%
 Error in routine  cdiaghg (251):
  problems computing cholesky
 %

I get this error even if I decrease the k-points to its minimum, and decrease 
the the cutoff energy to a ridiculously low value. I have to add I get this 
error for all major PPs.
So, I would really appreciate if somebody could give me any hint regarding this 
issue. In the meantime, I have attached the input file to this e-mail.


With kind regards,


Amin Mirzai,

PhD candidate

Dept. of Maskinteknologi,

Lund University,

P.O. Box 118,
SE-221 00 Lund,
Sweden



inputfile
Description: inputfile
___
users mailing list
users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users