Re: [Pw_forum] [QE-GPU] Bug regarding use of INTEL compilers in QE-GPU-5.4.0

[Paolo Giannozzi]

On Fri, Sep 22, 2017 at 7:30 PM, Phanikumar Pentyala  wrote:

In the previous thread (link) Prof. Paolo Giannozzi told that it's because
> of intel compilers problem and solution for this was linking of local blas.
>

“I really didn’t say everything I said”(Yogi Berra)

-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
___
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum

[Pw_forum] [QE-GPU] Bug regarding use of INTEL compilers in QE-GPU-5.4.0

[Phanikumar Pentyala]

Dear QE users

I am working on improving properties of doped metal oxides. Installation of
QE-GPU-5.4.0 was done without any errors then I am trying to optimize my
structure (CeO2). After successfully submission of my calculations, either
I am reaching to a stage with *oscillations of total energy *OR calculation
automatically *stops without any error**.* I thought of it's because of BAD
input parameters, so I changed so many parameters as suggested by so many
users in pw_forum like mixing_bata, mixing_dim, smearing, dgauss and based
on exchange correlation functional (NCPP, USPP) I changed ecutwfc and
ecutrho also, but NO use.

After through search, I got a similar previous reported problem (it's very
old): http://qe-forge.org/pipermail/pw_forum/2009-March/086558.html

Same above input file (mine) I tried in 4 processor CPU (desktop) installed
QE-5.4.0. Calculation achieved convergence smoothly.

In the previous thread (link) Prof. Paolo Giannozzi told that it's because
of intel compilers problem and solution for this was linking of local blas.
This I didn't understand, can you explain in elaborated way? These are my
server details

linux: *centOS*

Server: *FUJITSU Server PRIMERGY RX2540 M2*

GPU card:

*NVIDIA Tesla K40 (2 number)*


*intel 2017.1.132 for both mpi and mkl libraries*


*CUDA-8.0.44*
P.S: Input file attached

Thank You



Regards

Phanikumar
Research Scholar
IIT Kharagpur, Kharagpur
West Bengal, India.


ceo2.in
Description: Binary data
___
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum

Re: [Pw_forum] NMR Gipaw Calculations

[Davide Ceresoli]

Dear Alan,
 then probably 0.01 over 520 is ok. You might increase the
cutoff by 5-10 Ry, to improve the f-sum rule. Diagonalization='cg'
in gipaw, is usually faster.

Best,
 Davide


On 09/22/2017 04:11 PM, Ambrozio wrote:
> Dear Davide,
> 
> Thanks for replying,
> 
> I'm working with systems with more than 100 electrons. The systems have 130 
> or 
> more carbon atoms, it's expensive to me increase the number of k-points. I 
> need 
> optimize the computational cost, do you have a hint in this case?
> 
> Thanks a lot,
> 
> 
> Alan.
> 
> 2017-09-22 7:15 GMT-03:00 Davide Ceresoli  >:
> 
> Dear Alan,
>       I would try to reach an absolute error of the order of
> 0.001 on the f-sum rule, for systems up to ~100 electrons.
> I never investigated whether an absolute or relative error
> is the right choice.
> 
> Best regards,
>       Davide
> 
> 
> 
> On 09/21/2017 07:41 PM, Ambrozio wrote:
>  > Hi dear users and developers,
>  >
>  > I have a question regarding the chemical shielding convergence using
> Gipaw. As
>  > the best of my knowledge, the f-sum rule parameter is an indicative of 
> the
>  > convergence, corresponding to the number of valence electrons. I use a
> criterion
>  > of 0.01 for the absolute error in the f-sum rule tensor  principal
> components.
>  > I'd like to know if this criterion is correct? I mean ... should I 
> take the
>  > absolute error ? or the relative one?
>  >
>  > Regards,
>  >
>  > Alan,
>  >
>  > Federal University of Espirito Santo, Vitoria, Brazil.
>  >
>  > --
>  >
>  > Alan J. Romanel Ambrozio
>  > Bacharel em Física
>  > Mestre em Eng. de Materiais
>  > Doutorando em Física - PPGFis
>  >
>  >
> 
> ___
> Pw_forum mailing list
> Pw_forum@pwscf.org 
> http://pwscf.org/mailman/listinfo/pw_forum
> 
> 
> 
> 
> 
> -- 
> 
> Alan J. Romanel Ambrozio
> Bacharel em Física
> Mestre em Eng. de Materiais
> Doutorando em Física - PPGFis
> 
> 

-- 
+--+
   Davide Ceresoli
   CNR Institute of Molecular Science and Technology (CNR-ISTM)
   c/o University of Milan, via Golgi 19, 20133 Milan, Italy
   Email: davide.ceres...@cnr.it
   Phone: +39-02-50314276, +39-347-1001570 (mobile)
   Skype: dceresoli
   Website: http://sites.google.com/site/dceresoli/
+--+
___
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum

Re: [Pw_forum] NMR Gipaw Calculations

[Ambrozio]

Dear Davide,

Thanks for replying,

I'm working with systems with more than 100 electrons. The systems have 130
or more carbon atoms, it's expensive to me increase the number of k-points.
I need optimize the computational cost, do you have a hint in this case?

Thanks a lot,


Alan.

2017-09-22 7:15 GMT-03:00 Davide Ceresoli :

> Dear Alan,
>  I would try to reach an absolute error of the order of
> 0.001 on the f-sum rule, for systems up to ~100 electrons.
> I never investigated whether an absolute or relative error
> is the right choice.
>
> Best regards,
>  Davide
>
>
>
> On 09/21/2017 07:41 PM, Ambrozio wrote:
> > Hi dear users and developers,
> >
> > I have a question regarding the chemical shielding convergence using
> Gipaw. As
> > the best of my knowledge, the f-sum rule parameter is an indicative of
> the
> > convergence, corresponding to the number of valence electrons. I use a
> criterion
> > of 0.01 for the absolute error in the f-sum rule tensor  principal
> components.
> > I'd like to know if this criterion is correct? I mean ... should I take
> the
> > absolute error ? or the relative one?
> >
> > Regards,
> >
> > Alan,
> >
> > Federal University of Espirito Santo, Vitoria, Brazil.
> >
> > --
> >
> > Alan J. Romanel Ambrozio
> > Bacharel em Física
> > Mestre em Eng. de Materiais
> > Doutorando em Física - PPGFis
> >
> >
>
> ___
> Pw_forum mailing list
> Pw_forum@pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum




-- 

Alan J. Romanel Ambrozio
Bacharel em Física
Mestre em Eng. de Materiais
Doutorando em Física - PPGFis
___
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum

Re: [Pw_forum] Pw_forum Digest, Vol 122, Issue 22

[Aaditya Hambarde]

Dear Stefano,

We have used  in the BandCal.inp file as mentioned in the email. I
resimulated using QE 5.3 and was able to do the scf calculations but could
not proceed further with generating the bandstructure. What could be the
reason?

Best Wishes,
Aaditya (University of Michigan)

Best Regards,
Aaditya Hambarde
Phone: +91 98923 78138
Skype: aaditya.hambarde
Alternate Email: hamba...@umich.edu


On Fri, Sep 22, 2017 at 7:30 PM,  wrote:

> Send Pw_forum mailing list submissions to
> pw_forum@pwscf.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://pwscf.org/mailman/listinfo/pw_forum
> or, via email, send a message with subject or body 'help' to
> pw_forum-requ...@pwscf.org
>
> You can reach the person managing the list at
> pw_forum-ow...@pwscf.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Pw_forum digest..."
>
>
> Today's Topics:
>
>1. Re: Fwd: phonon calculation (Lorenzo Paulatto)
>2. NMR Gipaw Calculations (Ambrozio)
>3. Help With Bandstructure Calculation (Aaditya Hambarde)
>4. Re: Help With Bandstructure Calculation (Stefano de Gironcoli)
>5. Re: Wannier function (Andrea Ferretti)
>
>
> --
>
> Message: 1
> Date: Thu, 21 Sep 2017 12:08:19 +0200
> From: Lorenzo Paulatto 
> Subject: Re: [Pw_forum] Fwd: phonon calculation
> To: pw_forum@pwscf.org
> Message-ID: <9def3181-d613-6f36-d3e5-0750f0a9b...@impmc.upmc.fr>
> Content-Type: text/plain; charset=utf-8; format=flowed
>
> On 21/09/17 07:54, Suchitra Kumari wrote:
> > I am looking for phonon calculation for silicon nanowire using QE6.0.
> > Because of the complex structure and larger number of atoms (64), the
> > calculation is taking too much time and not converging till the end.
> > Will there be any extra parameters are to be incorporated for phonon
> > computations ?
>
>
> Dear Suchitra,
>
> 1. Do you manage to converge at least one q-point, or not at all?
> Because the first q-point Gamma=(0,0,0) is usually faster than the
> following ones (if you have any symmetry). If you do not manage to
> complete at least Gamma, the other ones will be worst.
>
> 2. If you do manage to complete one point, it is better to run them
> one-by-one, using start_q and last_q.
>
> 3. Do you really need to save the rotated charge density
> (drho_star%open=.true.)? This part can take some time and a lot of
> memory, but it is only used by third order calculation. A d3 calculation
> is prohibitive with 64 atoms anyway, in any case it is much longer than
> the phonon calculation.
>
> 4. Do you really need 8 q-points (I guess your nanotube is oriented
> along z)? You are assuming that there is significant ion-ion interaction
> up to 8 unit cells away. An 8x8x1 grid is largely sufficient for
> graphene, which only has a much smaller unit cell, with just 2 atoms.
>
>
>
> --
> Dr. Lorenzo Paulatto
> IdR @ IMPMC -- CNRS & Universit? Paris 6
> phone: +33 (0)1 442 79822 / skype: paulatz
> www:   http://www-int.impmc.upmc.fr/~paulatto/
> mail:  23-24/423 Bo?te courrier 115, 4 place Jussieu 75252 Paris C?dex 05
>
>
> --
>
> Message: 2
> Date: Thu, 21 Sep 2017 14:41:20 -0300
> From: Ambrozio 
> Subject: [Pw_forum] NMR Gipaw Calculations
> To: PWSCF Forum 
> Message-ID:
> 

Re: [Pw_forum] NMR Gipaw Calculations

[Davide Ceresoli]

Dear Alan,
 I would try to reach an absolute error of the order of
0.001 on the f-sum rule, for systems up to ~100 electrons.
I never investigated whether an absolute or relative error
is the right choice.

Best regards,
 Davide



On 09/21/2017 07:41 PM, Ambrozio wrote:
> Hi dear users and developers,
> 
> I have a question regarding the chemical shielding convergence using Gipaw. 
> As 
> the best of my knowledge, the f-sum rule parameter is an indicative of the 
> convergence, corresponding to the number of valence electrons. I use a 
> criterion 
> of 0.01 for the absolute error in the f-sum rule tensor  principal 
> components. 
> I'd like to know if this criterion is correct? I mean ... should I take the 
> absolute error ? or the relative one?
> 
> Regards,
> 
> Alan,
> 
> Federal University of Espirito Santo, Vitoria, Brazil.
> 
> -- 
> 
> Alan J. Romanel Ambrozio
> Bacharel em Física
> Mestre em Eng. de Materiais
> Doutorando em Física - PPGFis
> 
> 

___
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum

[Pw_forum] error during scf calculation

[Dhaval Satikunvar]

Dear all QE users

I am new in quantum espresso. During the calculation of cohesive energy of
Al, I am getting the following error. Calculation stops in few minutes.

Please help me to overcome this problem

"Al-EC.sh:   Line 12:  3194 Killed"

Thanks and regards,
Dhaval Satikunvar

Department of Physics,
Sardar Patel University,
Vallabh Vidyanagar.-388120.
Gujarat.
India.

Phone: 09426279465
___
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum

Re: [Pw_forum] Wannier function

[Andrea Ferretti]

Dear Mansoureh,

the conductor.x exectuable of WanT can perform the eigenchannel analysis of the 
transmittance but doesn't print anylonger the orbital-resolved contribution to 
the transmittance.

In case, it is not too difficult to extract the information from the code:
eg, around line 506 of want/src/transport/conductor.f90
the variable cond_aux(1:dimC) is filled with the quantities you are looking for 
(I think)

BTW: want-2.6.0-beta  is avaialable via the MaX userportal
http://userportal.max-centre.eu

hope it helps
Andrea

> Dear all
> I would like find s, px, py and pz contribution in I-V curve for a system, 
> separately. How can I set wannier function in WanT code?
> Regards
> Mansoureh Pashangpour
> IIAU
> Tehran,Iran
> 
> Sent from my iPhone
> ___
> Pw_forum mailing list
> Pw_forum@pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>

-- 
Andrea Ferretti, PhD
S3 Center, Istituto Nanoscienze, CNR
via Campi 213/A, 41125, Modena, Italy
Tel: +39 059 2055322;  Skype: andrea_ferretti
URL: http://www.nano.cnr.it

___
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum