Re: [QE-users] Are my PBE0 calculations running correctly?

[Giuseppe Mattioli]

I want to do a comparison between GGA, GGA+U and PBE0. In GGA and  
GGA+U I used PAW pp, I don`t know if it will be right to use another  
type of pseudopotential, and be able to compare them because in  
previous publications I noticed that they always use same  
pseudopotential.


I think you need minutes to perform a new set of GGA and DFT+U  
calculations using NC PPs. Otherwise you might lose years to perform  
PBE0 calculations using PAW.


Where can I find the manual please. Is ecutfock a parameter that I  
should include in the input file.


Yes, it is. And the old good textlike manual is always here:

/your_path_to_QE_main_dir/Doc/INPUT_PW.txt

Do I have to change something in the input file as a consequence of  
changing ibrav from 0 to 1


A few things:

substitute

  ibrav = 1

 and

  A =6.05844*1.005470117

and remove

CELL_PARAMETERS {alat}
   1.005470117   0.0   0.0
   0.0   1.005470117   0.0
   0.0   0.0   1.005470117

There are also a few things you don't need, like

tprnfor=.true.,
tstress=.true.,
etot_conv_thr =1e-8
forc_conv_thr =1e-7

diago_david_ndim = 4

and a few things that are not wrong but I would use with different  
values, like


conv_thr=1d-07,
mixing_beta=0.1d0,

HTH
Giuseppe

Quoting NAIMI SALMA :


Dear GIUSEPPE,

--->First of all it is not ZnO but CuI :-D
I'm working on CuI but I got confused and commented  in the post of  
another user . I receveid a mail containing her post. And I typed  
her name too in my comment. Sorry but I'm still not familliar with  
the use of the forum, as I'm new here :)


>Please *don't use PAW pseudopotentials for EXX calculations*! There is
no gain and it makes the EXX part of the calculation less stable. Use
norm-conserving pseudopotentials instead.

I want to do a comparison between GGA, GGA+U and PBE0. In GGA and  
GGA+U I used PAW pp, I don`t know if it will be right to use another  
type of pseudopotential, and be able to compare them because in  
previous publications I noticed that they always use same  
pseudopotential.


> Then find in the manual the variable ecutfock and reduce the cutoff
for the calculation of EXX to something between ecutwfc and 2*ecutwfc.
You save a lot of time without losing accuracy.

Where can I find the manual please. Is ecutfock a parameter that I  
should include in the input file.



> Finally, your cell is cubic. Use ibrav=1 instead of 0, as this can
reduce the k-points actually used by the code in the automatic grid.

Do I have to change something in the input file as a consequence of  
changing ibrav from 0 to 1.












Salma NAIMI
PhD student-Mohamed 5 University-Rabat

From: users  on behalf of  
Giuseppe Mattioli 

Sent: Thursday, March 2, 2023 12:46 PM
To: Quantum ESPRESSO users Forum 
Subject: Re: [QE-users] Are my PBE0 calculations running correctly?


First of all it is not ZnO but CuI :-D

Please *don't use PAW pseudopotentials for EXX calculations*! There is
no gain and it makes the EXX part of the calculation less stable. Use
norm-conserving pseudopotentials instead.

Then find in the manual the variable ecutfock and reduce the cutoff
for the calculation of EXX to something between ecutwfc and 2*ecutwfc.
You save a lot of time without losing accuracy.

Finally, your cell is cubic. Use ibrav=1 instead of 0, as this can
reduce the k-points actually used by the code in the automatic grid.

HTH
Giuseppe

Quoting NAIMI SALMA :


Dear GIUSEPPE,

It's my first time to use PBE0.

I attached my input and output files.






--
Salma NAIMI
PhD student-Mohamed 5 University-Rabat


From: users  on behalf of
Giuseppe Mattioli 
Sent: Thursday, March 2, 2023 12:10 PM
To: Quantum ESPRESSO users Forum 
Subject: Re: [QE-users] Are my PBE0 calculations running correctly?


It is a system error, not a code error. It might be an out of memory
error. Keep in mind that in the first scf iteration you calculate
usual GGA orbitals/ks_potentials and only here


 Using ACE for calculation of exact exchange


you calculate for the first time EXX. Close to the beginning of the
output there is an estimate of memory usage, something like this:

  number of k points= 2  Gaussian smearing, width (Ry)=  0.0100
cart. coord. in units 2pi/alat
 k(1) = (   0.000   0.000   0.000), wk =   1.000
 k(2) = (  -0.500   0.000   0.000), wk =   1.000

  Dense  grid:  6118035 G-vectors FFT dimensions: (  72, 512, 360)

  Estimated max dynamical RAM per process >   1.35 GB

  Estimated total dynamical RAM >  43.17 GB

Is it compatible with the resources available for your desktop pc? Can
you send your input/output files?
HTH
Giuseppe

Quoting NAIMI SALMA :


Dear experts,

I did a PBE0 for my material. and after some time I got this message
in the terminal:



Invalid 

Re: [QE-users] I'm confused about how to use PBE0

[Battal Malika]

Thank you Giuseppe those are useful informations, I will search for the video 
first.














--
Malika battal
student - Qatar University


De : users  de la part de NAIMI SALMA 

Envoyé : jeudi 2 mars 2023 10:38
À : Quantum ESPRESSO users Forum 
Objet : Re: [QE-users] I'm confused about how to use PBE0

Dear all,

Thank you for the valuable informations.

I ran the calculations and after some time i get the following message in 
terminal:


Invalid MIT-MAGIC-COOKIE-1 key[-Precision-3640-Tower:198424] Read -1, expected 
68400, errno = 3
--
Primary job  terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.
--
--
mpirun noticed that process rank 0 with PID 0 on node -Precision-3640-Tower 
exited on signal 9 (Killed).
--

[1]+  Exit 137


And when I go to the last lines in the output file , I find this:

 convergence has been achieved in   8 iterations

 Using ACE for calculation of exact exchange

 EXX grid:   175829 G-vectors FFT dimensions: (  72,  72,  72)



It seems like I'm facing a problem. How can I fix it, please?





-
Malika battal
student - Qatar University


From: users  on behalf of Giuseppe 
Mattioli 
Sent: Thursday, March 2, 2023 10:46 AM
To: Quantum ESPRESSO users Forum 
Subject: Re: [QE-users] I'm confused about how to use PBE0


A 4-atom unit cell of ZnO can likely run on a desktop/laptop pc using
PBE0, as long as you don't care too much about your hardware
(desktop/laptop pc are generally not happy if you stress them for
hours with codes designed mainly for workstation/HPC use). You can
still find somewhere in the internet a famous video posted many years
ago by prof. Marzari who run a 2-atom Si cell on his Nokia N-something
cell phone. :-D

Completely different is the case of a ZnO bulk supercell or even
surface slab. No way in this case, you need suitable resources.

Anyway, ZnO is a nasty beast, as you can have spurious mixing of Zn 3d
and O 2p shells. If you are mainly interested in recovering a band gap
close to the optically measured one, then you should mess with the EXX
fraction of PBE0 (or HSE), raising it to 35-40%.

HTH
Giuseppe

Quoting Battal Malika :

> Dear Lorenzo,
>
> Sorry it's not a laptop but a desktop computer, would it be possible
> in this case?
>
>
>
>
>
>
>
>
>
>
>
> 
> Malika battal
> student - Qatar University
>
>
>
>
>
>
>
> 
> De : users  de la part de
> Lorenzo Paulatto 
> Envoyé : mercredi 1 mars 2023 22:02
> À : users@lists.quantum-espresso.org 
> Objet : Re: [QE-users] I'm confused about how to use PBE0
>
>
> Dear Malika,
>
> the Fock exchange (also known as exact exchange) included in PBE0 is
> very computationally intensive, it is unlikely that you'll be able
> to run it on a laptop.
>
> kind regards
>
>
> On 01/03/2023 22:30, Battal Malika wrote:
> Hello,
> I noticed that bandgap of ZnO is underestimated when using GGA. It
> seems like PBE0 is the most accurate methode in my case to find a
> value close to the experimental one. Neverthless I don't know how to
> use it and how much time i'll need to run it in a laptop with 8
> processors.
> Can you please clarify my doubts?
>
>
>
>
>
> 
> Malika battal
> student - Qatar University
>
>
>
> ___
> 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
> ([http://www.max-centre.eu)
> users mailing list
> users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
>
> --
> Dr. Lorenzo Paulatto
> IdR @ IMPMC - CNRS UMR 7590 & Sorbonne Université
> phone: +33 (0)1 442 79822 / skype: paulatz
> 

Re: [QE-users] Are my PBE0 calculations running correctly?

[Giuseppe Mattioli]

First of all it is not ZnO but CuI :-D

Please *don't use PAW pseudopotentials for EXX calculations*! There is  
no gain and it makes the EXX part of the calculation less stable. Use  
norm-conserving pseudopotentials instead.


Then find in the manual the variable ecutfock and reduce the cutoff  
for the calculation of EXX to something between ecutwfc and 2*ecutwfc.  
You save a lot of time without losing accuracy.


Finally, your cell is cubic. Use ibrav=1 instead of 0, as this can  
reduce the k-points actually used by the code in the automatic grid.


HTH
Giuseppe

Quoting NAIMI SALMA :


Dear GIUSEPPE,

It's my first time to use PBE0.

I attached my input and output files.






--
Salma NAIMI
PhD student-Mohamed 5 University-Rabat


From: users  on behalf of  
Giuseppe Mattioli 

Sent: Thursday, March 2, 2023 12:10 PM
To: Quantum ESPRESSO users Forum 
Subject: Re: [QE-users] Are my PBE0 calculations running correctly?


It is a system error, not a code error. It might be an out of memory
error. Keep in mind that in the first scf iteration you calculate
usual GGA orbitals/ks_potentials and only here


 Using ACE for calculation of exact exchange


you calculate for the first time EXX. Close to the beginning of the
output there is an estimate of memory usage, something like this:

  number of k points= 2  Gaussian smearing, width (Ry)=  0.0100
cart. coord. in units 2pi/alat
 k(1) = (   0.000   0.000   0.000), wk =   1.000
 k(2) = (  -0.500   0.000   0.000), wk =   1.000

  Dense  grid:  6118035 G-vectors FFT dimensions: (  72, 512, 360)

  Estimated max dynamical RAM per process >   1.35 GB

  Estimated total dynamical RAM >  43.17 GB

Is it compatible with the resources available for your desktop pc? Can
you send your input/output files?
HTH
Giuseppe

Quoting NAIMI SALMA :


Dear experts,

I did a PBE0 for my material. and after some time I got this message
in the terminal:



Invalid MIT-MAGIC-COOKIE-1 key[-Precision-3640-Tower:198424] Read
-1, expected 68400, errno = 3
--
Primary job  terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.
--
--
mpirun noticed that process rank 0 with PID 0 on node
-Precision-3640-Tower exited on signal 9 (Killed).
--

[1]+  Exit 137




And when I go to the last lines in the output file , I find this:

 convergence has been achieved in   8 iterations

 Using ACE for calculation of exact exchange

 EXX grid:   175829 G-vectors FFT dimensions: (  72,  72,  72)



It seems like I'm facing a problem. How can I fix it, please?

For the PBE0 I added the following lines to my input file in the
 block:


input_dft='pbe0',
nqx1 = 1, nqx2 =1, nqx3 =1,
x_gamma_extrapolation = .true.
exxdiv_treatment = 'gygi-baldereschi'









---

Salma NAIMI




PhD student-Mohamed 5 University-Rabat




GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: 

___
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




GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: 

___
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] Are my PBE0 calculations running correctly?

[Giuseppe Mattioli]

It is a system error, not a code error. It might be an out of memory  
error. Keep in mind that in the first scf iteration you calculate  
usual GGA orbitals/ks_potentials and only here



 Using ACE for calculation of exact exchange


you calculate for the first time EXX. Close to the beginning of the  
output there is an estimate of memory usage, something like this:


 number of k points= 2  Gaussian smearing, width (Ry)=  0.0100
   cart. coord. in units 2pi/alat
k(1) = (   0.000   0.000   0.000), wk =   1.000
k(2) = (  -0.500   0.000   0.000), wk =   1.000

 Dense  grid:  6118035 G-vectors FFT dimensions: (  72, 512, 360)

 Estimated max dynamical RAM per process >   1.35 GB

 Estimated total dynamical RAM >  43.17 GB

Is it compatible with the resources available for your desktop pc? Can  
you send your input/output files?

HTH
Giuseppe

Quoting NAIMI SALMA :


Dear experts,

I did a PBE0 for my material. and after some time I got this message  
in the terminal:




Invalid MIT-MAGIC-COOKIE-1 key[-Precision-3640-Tower:198424] Read  
-1, expected 68400, errno = 3

--
Primary job  terminated normally, but 1 process returned
a non-zero exit code. Per user-direction, the job has been aborted.
--
--
mpirun noticed that process rank 0 with PID 0 on node  
-Precision-3640-Tower exited on signal 9 (Killed).

--

[1]+  Exit 137




And when I go to the last lines in the output file , I find this:

 convergence has been achieved in   8 iterations

 Using ACE for calculation of exact exchange

 EXX grid:   175829 G-vectors FFT dimensions: (  72,  72,  72)



It seems like I'm facing a problem. How can I fix it, please?

For the PBE0 I added the following lines to my input file in the  
 block:



input_dft='pbe0',
nqx1 = 1, nqx2 =1, nqx3 =1,
x_gamma_extrapolation = .true.
exxdiv_treatment = 'gygi-baldereschi'









---

Salma NAIMI




PhD student-Mohamed 5 University-Rabat




GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: 

___
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] I'm confused about how to use PBE0

[Giuseppe Mattioli]

A 4-atom unit cell of ZnO can likely run on a desktop/laptop pc using  
PBE0, as long as you don't care too much about your hardware  
(desktop/laptop pc are generally not happy if you stress them for  
hours with codes designed mainly for workstation/HPC use). You can  
still find somewhere in the internet a famous video posted many years  
ago by prof. Marzari who run a 2-atom Si cell on his Nokia N-something  
cell phone. :-D


Completely different is the case of a ZnO bulk supercell or even  
surface slab. No way in this case, you need suitable resources.


Anyway, ZnO is a nasty beast, as you can have spurious mixing of Zn 3d  
and O 2p shells. If you are mainly interested in recovering a band gap  
close to the optically measured one, then you should mess with the EXX  
fraction of PBE0 (or HSE), raising it to 35-40%.


HTH
Giuseppe

Quoting Battal Malika :


Dear Lorenzo,

Sorry it's not a laptop but a desktop computer, would it be possible  
in this case?













Malika battal
student - Qatar University








De : users  de la part de  
Lorenzo Paulatto 

Envoyé : mercredi 1 mars 2023 22:02
À : users@lists.quantum-espresso.org 
Objet : Re: [QE-users] I'm confused about how to use PBE0


Dear Malika,

the Fock exchange (also known as exact exchange) included in PBE0 is  
very computationally intensive, it is unlikely that you'll be able  
to run it on a laptop.


kind regards


On 01/03/2023 22:30, Battal Malika wrote:
Hello,
I noticed that bandgap of ZnO is underestimated when using GGA. It  
seems like PBE0 is the most accurate methode in my case to find a  
value close to the experimental one. Neverthless I don't know how to  
use it and how much time i'll need to run it in a laptop with 8  
processors.

Can you please clarify my doubts?






Malika battal
student - Qatar University



___
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

--
Dr. Lorenzo Paulatto
IdR @ IMPMC - CNRS UMR 7590 & Sorbonne Université
phone: +33 (0)1 442 79822 / skype: paulatz
http://www.impmc.upmc.fr/~paulatto/ - https://anharmonic.github.io/
23-24/423 B115, 4 place Jussieu 75252 Paris CX 05




GIUSEPPE MATTIOLI
CNR - ISTITUTO DI STRUTTURA DELLA MATERIA
Via Salaria Km 29,300 - C.P. 10
I-00015 - Monterotondo Scalo (RM)
Mob (*preferred*) +39 373 7305625
Tel + 39 06 90672342 - Fax +39 06 90672316
E-mail: 

___
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] I'm confused about how to use PBE0

[Stefano Baroni]

Experience is the Mother of Science. Just try! SB

On 2 Mar 2023, at 09:39, Battal Malika  wrote:

Dear Lorenzo,

Sorry it's not a laptop but a desktop computer, would it be possible in this 
case?












Malika battal
student - Qatar University








De : users 
mailto:users-boun...@lists.quantum-espresso.org>>
 de la part de Lorenzo Paulatto 
mailto:lorenzo.paula...@cnrs.fr>>
Envoyé : mercredi 1 mars 2023 22:02
À : users@lists.quantum-espresso.org 
mailto:users@lists.quantum-espresso.org>>
Objet : Re: [QE-users] I'm confused about how to use PBE0

Dear Malika,
the Fock exchange (also known as exact exchange) included in PBE0 is very 
computationally intensive, it is unlikely that you'll be able to run it on a 
laptop.
kind regards

On 01/03/2023 22:30, Battal Malika wrote:
Hello,
I noticed that bandgap of ZnO is underestimated when using GGA. It seems like 
PBE0 is the most accurate methode in my case to find a value close to the 
experimental one. Neverthless I don't know how to use it and how much time i'll 
need to run it in a laptop with 8 processors.
Can you please clarify my doubts?






Malika battal
student - Qatar University


___
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

--
Dr. Lorenzo Paulatto
IdR @ IMPMC - CNRS UMR 7590 & Sorbonne Université
phone: +33 (0)1 442 79822 / skype: paulatz
http://www.impmc.upmc.fr/~paulatto/ - https://anharmonic.github.io/
23-24/423 B115, 4 place Jussieu 75252 Paris CX 05
___
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

—
Stefano Baroni - Trieste —  http://stefano.baroni.me




___
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] I'm confused about how to use PBE0

[Battal Malika]

Dear Lorenzo,

Sorry it's not a laptop but a desktop computer, would it be possible in this 
case?












Malika battal
student - Qatar University








De : users  de la part de Lorenzo 
Paulatto 
Envoyé : mercredi 1 mars 2023 22:02
À : users@lists.quantum-espresso.org 
Objet : Re: [QE-users] I'm confused about how to use PBE0


Dear Malika,

the Fock exchange (also known as exact exchange) included in PBE0 is very 
computationally intensive, it is unlikely that you'll be able to run it on a 
laptop.

kind regards


On 01/03/2023 22:30, Battal Malika wrote:
Hello,
I noticed that bandgap of ZnO is underestimated when using GGA. It seems like 
PBE0 is the most accurate methode in my case to find a value close to the 
experimental one. Neverthless I don't know how to use it and how much time i'll 
need to run it in a laptop with 8 processors.
Can you please clarify my doubts?






Malika battal
student - Qatar University



___
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

--
Dr. Lorenzo Paulatto
IdR @ IMPMC - CNRS UMR 7590 & Sorbonne Université
phone: +33 (0)1 442 79822 / skype: paulatz
http://www.impmc.upmc.fr/~paulatto/ - https://anharmonic.github.io/
23-24/423 B115, 4 place Jussieu 75252 Paris CX 05
___
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