Re: [Wien] ** testerror: Error in Parallel LAPW

[Gavin Abo]

The "Host key verification failed" is an error from ssh [1].


Thus, it seems like you need to fix your ssh so that WIEN2k can connect 
to your remote node (in your case below, it looks like the remote node 
is lxbk1177).



It looks like there is an ssh example on slide 10 of the WIEN2k 
presentation at [2].



I believe it is common to use ssh-keygen to create a key pair (private 
and public key) on the head node and then use ssh-copy-id to put the 
public key on each of the remote nodes.  However, ssh can be uniquely 
configured for a computer system.  So, you might want to search online 
for different examples on how ssh has been configured.  One example that 
might be helpful should be at [3].



[1] 
https://askubuntu.com/questions/45679/ssh-connection-problem-with-host-key-verification-failed-error


[2] 
https://www.bc.edu/content/dam/bc1/schools/mcas/physics/pdf/wien2k/PB-installation.pdf


[3] 
https://www.digitalocean.com/community/tutorials/ssh-essentials-working-with-ssh-servers-clients-and-keys



Kind Regards,

Gavin

WIEN2k user


On 6/20/2023 3:25 PM, Ilias Miroslav, doc. RNDr., PhD. wrote:

Dear Professor Blaha,

thanks, I used PATH variable extension instead of linking;

it  crashed with the message  "Host key verification failed. "

Herethe content of file 
/lustre/ukt/milias/scratch/Wien2k_23.2_job.main.N1.n4.jid3009460/LvO2onQg/.machines:

1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177

Job is running on lxbk1177, with 8 cpus allocated;

and this is from log :

running x dstart :
starting parallel dstart at Tue 20 Jun 2023 05:16:21 PM CEST
 .machine0 : processors
running dstart in single mode
STOP DSTART ENDS
10.249u 0.322s 0:11.19 94.3%    0+0k 158496+101160io 437pf+0w

running 'run_lapw -p -ec 0.0001 -NI'
STOP  LAPW0 END
Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def ;fixerr
or_lapw ${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdout1_$loop; 
if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw .stdout1_$loop > .
temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; grep -v \% 
.temp1_$loop | perl -e "print stderr " )

Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def 
;fixerror_lapw ${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw 
.stdout1_$loop > .temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; 
grep -v \% .temp1_$loop | perl -e "print stderr " )

Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def 
;fixerror_lapw ${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw 
.stdout1_$loop > .temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; 
grep -v \% .temp1_$loop | perl -e "print stderr " )

Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def 
;fixerror_lapw ${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw 
.stdout1_$loop > .temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; 
grep -v \% .temp1_$loop | perl -e "print stderr " )

Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def 
;fixerror_lapw ${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw 
.stdout1_$loop > .temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; 
grep -v \% .temp1_$loop | perl -e "print stderr " )

Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def 
;fixerror_lapw ${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw 
.stdout1_$loop > .temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; 
grep -v \% .temp1_$loop | perl -e "print stderr " )

Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def 
;fixerror_lapw ${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw 
.stdout1_$loop > .temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; 
grep -v \% .temp1_$loop | perl -e "print stderr " )

Host key verification failed.
[1]    Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def 
;fixerror_lapw ${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw 
.stdout1_$loop > .temp1_$loop; grep \% 

[Wien] Difference in DOS and BAND graphics

[Hülya Gürçay]

Dear WIEN2k users,

I made optimisation for MnVZrP , i found the equilibrium lattice parameter
6.07 A,
I used this lattice parameter in SCF calculation,
I copied this SCF file to a new folder and plotted Band and DOS graphs
through the interface.
In the band graph, there is a gap in the spin down channel and the material
is semi-metal, while in the DOS graph, metal appears in both the spin up
and spin down channels.
I both plotted in eV and Ry units,

How can i fix this incompatibility?

XC: GGA-PBE
K points: 20,20,20
Rkmax:8; Lmax:12,
Cc: 0.0001; Ec:0.1

This material has been calculated before with different code , here
https://pubs.rsc.org/en/content/articlelanding/2020/ra/d0ra04633g

Sincerely,
Hülya
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] ** testerror: Error in Parallel LAPW

[Ilias Miroslav, doc. RNDr., PhD.]

Dear Professor Blaha,

thanks, I used PATH variable extension instead of linking;

it  crashed with the message  "Host key verification failed. "

Here the content of file 
/lustre/ukt/milias/scratch/Wien2k_23.2_job.main.N1.n4.jid3009460/LvO2onQg/.machines:
1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177
1:lxbk1177

Job is running on lxbk1177, with 8 cpus allocated;

and this is from log :

running x dstart :
starting parallel dstart at Tue 20 Jun 2023 05:16:21 PM CEST
 .machine0 : processors
running dstart in single mode
STOP DSTART ENDS
10.249u 0.322s 0:11.19 94.3%0+0k 158496+101160io 437pf+0w

running 'run_lapw -p -ec 0.0001 -NI'
STOP  LAPW0 END
Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def ;fixerr
or_lapw ${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdout1_$loop; if ( -f 
.stdout1_$loop ) bashtime2csh.pl_lapw .stdout1_$loop > .
temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; grep -v \% .temp1_$loop | 
perl -e "print stderr " )
Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def ;fixerror_lapw 
${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw .stdout1_$loop > 
.temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; grep -v \% .temp1_$loop | 
perl -e "print stderr " )
Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def ;fixerror_lapw 
${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw .stdout1_$loop > 
.temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; grep -v \% .temp1_$loop | 
perl -e "print stderr " )
Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def ;fixerror_lapw 
${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw .stdout1_$loop > 
.temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; grep -v \% .temp1_$loop | 
perl -e "print stderr " )
Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def ;fixerror_lapw 
${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw .stdout1_$loop > 
.temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; grep -v \% .temp1_$loop | 
perl -e "print stderr " )
Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def ;fixerror_lapw 
${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw .stdout1_$loop > 
.temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; grep -v \% .temp1_$loop | 
perl -e "print stderr " )
Host key verification failed.
[1]  + Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def ;fixerror_lapw 
${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw .stdout1_$loop > 
.temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; grep -v \% .temp1_$loop | 
perl -e "print stderr " )
Host key verification failed.
[1]Done  ( ( $remote $machine[$p] "cd 
$PWD;$set_OMP_NUM_THREADS;$t $taskset0 $exe ${def}_$loop.def ;fixerror_lapw 
${def}_$loop"; rm -f .lock_$lockfile[$p] ) >& .stdo
ut1_$loop; if ( -f .stdout1_$loop ) bashtime2csh.pl_lapw .stdout1_$loop > 
.temp1_$loop; grep \% .temp1_$loop >> .time1_$loop; grep -v \% .temp1_$loop | 
perl -e "print stderr " )
LvO2onQg.scf1_1: No such file or directory.
grep: *scf1*: No such file or directory
STOP FERMI - Error
cp: cannot stat '.in.tmp': No such file or directory
grep: *scf1*: No such file or directory

>   stop error



file ":parallel"

starting parallel lapw1 at Tue 20 Jun 2023 05:17:08 PM CEST
lxbk1177(4)  lxbk1177(3)  lxbk1177(3)  lxbk1177(3)  
lxbk1177(3)  lxbk1177(3)  lxbk1177(3)  l
xbk1177(3)Summary of lapw1para:
  lxbk1177  k=25user=0  wallclock=0
<-  done at Tue 20 Jun 2023 05:17:14 PM CEST
-
->  starting Fermi on lxbk1177 at Tue 20 Jun 2023 05:17:15 PM CEST
**  LAPW2 crashed at Tue 20 Jun 2023 05:17:16 PM CEST
**  check ERROR FILES!
-




___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  

Re: [Wien] calculation of surface/slab using lmbj potential

[Peter Blaha]

But then you need also to include   -nlvdw   in run_lapw, (also in the 
previous bulk optimization.



Am 20.06.2023 um 12:08 schrieb Burhan Ahmed:

To include van der waals interaction (van der calculation)

On Tue, 20 Jun, 2023, 3:31 pm Peter Blaha, > wrote:


Looks ok, beside:

Why do you do     Init_nl_vdw     ???

Am 20.06.2023 um 11:31 schrieb Burhan Ahmed:
 > Dear experts, I am having a compact cluster with 100gb RAM, 36
cores and
 > 12 TB SATA hard disk with CentOS.
 >
 > I want to calculate the band structure by considering the van der
waals
 > interaction of 6ql topological insulator. Below are the steps I have
 > used for the detailed calculation.
 >
 > Optimized the cell with PBE.
 >
 > After initialization, I run a pbe scf using
 >
 > “run_lapw -p -ec 0.0001 -cc 0.001 -NI”
 >
 > Save the result using
 >
 > “save_lapw *pbe*”
 >
 > In the next step
 >
 > “init_so_lapw” ( took the default parameters, magnetization is 0 0 1)
 >
 > “Init_nl_vdw “ (kernel type 1)
 >
 > “init_mbj_lapw”
 >
 > “run_lapw -p -so -i 1 -NI”
 >
 > “save_lapw *mbj*”
 >
 > “init_mbj_lapw”
 >
 > Took the selection 2 for surface calculation, and lmbj for potential
 >
 > And finally I run “run_lapw -p -lmbj -so -ec 0.0001 -cc 0.001 -i
999 -NI”
 >
 > And then I save the result using save_lapw *lmbj*.
 >
 > I want to know whether these are the correct steps for
calculating band
 > structure with lmbj potential available in wien2k_23.
 >
 > Thanks in Advance
 >
 > Regards
 >
 > Burhan Ahmed
 >
 > *Research Scholar, AUS *
 >
 >
 > ___
 > Wien mailing list
 > Wien@zeus.theochem.tuwien.ac.at

 > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien

 > SEARCH the MAILING-LIST at:
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html 


-- 
--

Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300
Email: peter.bl...@tuwien.ac.at    
WIEN2k: http://www.wien2k.at 

WWW: http://www.imc.tuwien.ac.at 
-
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at 
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien

SEARCH the MAILING-LIST at:
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html 



___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html


--
--
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300
Email: peter.bl...@tuwien.ac.atWIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at
-
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] ** testerror: Error in Parallel LAPW

[pluto via Wien]

Dear Miro,

It is hard to give your a meaningful answer with little info, but I will 
try my best guess because I needed to set this up recently. I assume 
that you want to use k-parallel and you don't have mpi.


With a serial job you automatically run on a single node. Single node is 
a physical computer with a physical CPU, but typically with 4 memory 
channels to it can run 8 jobs in parallel.


With k-parallel you need to define nodes on which k-points are 
calculated. With slurm, maybe things will work if you create 8 
"localhost" lines in .machines file, because this will still run on a 
single node that is assigned automatically. But things probably won't 
work if you create lines such as "node001", "node002" etc (depending on 
the names of the nodes in your cluster). And to take an advantage of the 
cluster you need to use as many nodes as possible.


Now the problem is, that k-parallel works assuming you can ssh to every 
node without a password. This is typically forbidden in the slurm 
environment. Prof. Blaha provides workarounds, but to me their 
implementation seems complicated (I not an expert): 
http://www.wien2k.at/reg_user/faq/pbs.html


I am using an older cluster where it is possible to allocate nodes, and 
with this allocation comes automatically passwordless ssh to these 
nodes. Then the slurm workarounds are not needed. Maybe you can talk to 
your administrator if this is possible in your cluster, because I think 
typically this is blocked.


Best,
Lukasz





On 2023-06-20 10:18, Ilias Miroslav, doc. RNDr., PhD. wrote:

Hello,

 I am able to run serial SCF via SLURM


https://github.com/miroi/open-collection/blob/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg/virgo_slurm_wien2kgnupar_fromdstart.01


 but when trying parallel

https://github.com/miroi/open-collection/blob/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg/virgo_slurm_wien2kgnupar_fromdstart.02


 I get lapw2.error

 'LAPW2' - can't open unit: 30

'LAPW2' -filename: LvO2onQg.energy_1

**  testerror: Error in Parallel LAPW2

 The file "LvO2onQg.energy" is correct in serial mode.

 Seems that LvO2onQg.energy_1 file is not produces in parallel run ?

 All files are
https://github.com/miroi/open-collection/tree/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg
[1]

 Best,

 Miro


Links:
--
[1]
https://github.com/miroi/open-collection/tree/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] calculation of surface/slab using lmbj potential

[Burhan Ahmed]

To include van der waals interaction (van der calculation)

On Tue, 20 Jun, 2023, 3:31 pm Peter Blaha,  wrote:

> Looks ok, beside:
>
> Why do you do Init_nl_vdw ???
>
> Am 20.06.2023 um 11:31 schrieb Burhan Ahmed:
> > Dear experts, I am having a compact cluster with 100gb RAM, 36 cores and
> > 12 TB SATA hard disk with CentOS.
> >
> > I want to calculate the band structure by considering the van der waals
> > interaction of 6ql topological insulator. Below are the steps I have
> > used for the detailed calculation.
> >
> > Optimized the cell with PBE.
> >
> > After initialization, I run a pbe scf using
> >
> > “run_lapw -p -ec 0.0001 -cc 0.001 -NI”
> >
> > Save the result using
> >
> > “save_lapw *pbe*”
> >
> > In the next step
> >
> > “init_so_lapw” ( took the default parameters, magnetization is 0 0 1)
> >
> > “Init_nl_vdw “ (kernel type 1)
> >
> > “init_mbj_lapw”
> >
> > “run_lapw -p -so -i 1 -NI”
> >
> > “save_lapw *mbj*”
> >
> > “init_mbj_lapw”
> >
> > Took the selection 2 for surface calculation, and lmbj for potential
> >
> > And finally I run “run_lapw -p -lmbj -so -ec 0.0001 -cc 0.001 -i 999 -NI”
> >
> > And then I save the result using save_lapw *lmbj*.
> >
> > I want to know whether these are the correct steps for calculating band
> > structure with lmbj potential available in wien2k_23.
> >
> > Thanks in Advance
> >
> > Regards
> >
> > Burhan Ahmed
> >
> > *Research Scholar, AUS *
> >
> >
> > ___
> > Wien mailing list
> > Wien@zeus.theochem.tuwien.ac.at
> > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> > SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>
> --
> --
> Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
> Phone: +43-1-58801-165300
> Email: peter.bl...@tuwien.ac.atWIEN2k: http://www.wien2k.at
> WWW:   http://www.imc.tuwien.ac.at
> -
> ___
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] calculation of surface/slab using lmbj potential

[Peter Blaha]

Looks ok, beside:

Why do you do Init_nl_vdw ???

Am 20.06.2023 um 11:31 schrieb Burhan Ahmed:
Dear experts, I am having a compact cluster with 100gb RAM, 36 cores and 
12 TB SATA hard disk with CentOS.


I want to calculate the band structure by considering the van der waals 
interaction of 6ql topological insulator. Below are the steps I have 
used for the detailed calculation.


Optimized the cell with PBE.

After initialization, I run a pbe scf using

“run_lapw -p -ec 0.0001 -cc 0.001 -NI”

Save the result using

“save_lapw *pbe*”

In the next step

“init_so_lapw” ( took the default parameters, magnetization is 0 0 1)

“Init_nl_vdw “ (kernel type 1)

“init_mbj_lapw”

“run_lapw -p -so -i 1 -NI”

“save_lapw *mbj*”

“init_mbj_lapw”

Took the selection 2 for surface calculation, and lmbj for potential

And finally I run “run_lapw -p -lmbj -so -ec 0.0001 -cc 0.001 -i 999 -NI”

And then I save the result using save_lapw *lmbj*.

I want to know whether these are the correct steps for calculating band 
structure with lmbj potential available in wien2k_23.


Thanks in Advance

Regards

Burhan Ahmed

*Research Scholar, AUS *


___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html


--
--
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300
Email: peter.bl...@tuwien.ac.atWIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at
-
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

[Wien] calculation of surface/slab using lmbj potential

[Burhan Ahmed]

Dear experts, I am having a compact cluster with 100gb RAM, 36 cores and 12 TB SATA hard disk with CentOS.I want to calculate the band structure by considering the van der waals interaction of 6ql topological insulator. Below are the steps I have used for the detailed calculation.Optimized the cell with PBE.After initialization, I run a pbe scf using “run_lapw -p -ec 0.0001 -cc 0.001 -NI”Save the result using “save_lapw pbe”In the next step “init_so_lapw” ( took the default parameters, magnetization is 0 0 1)“Init_nl_vdw “ (kernel type 1)“init_mbj_lapw”“run_lapw -p -so -i 1 -NI”“save_lapw mbj”“init_mbj_lapw”Took the selection 2 for surface calculation, and lmbj for potential And finally I run “run_lapw -p -lmbj -so -ec 0.0001 -cc 0.001 -i 999 -NI”And then I save the result using save_lapw lmbj. I want to know whether these are the correct steps for calculating band structure with lmbj potential available in wien2k_23. Thanks in Advance   RegardsBurhan AhmedResearch Scholar, AUS  
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] Calculation of surface state using lmbj potential with SO coupling.

[Peter Blaha]

DFT-D3 is a method for van der Waals interactions, which are not well 
described by GGA or mBJ. It has nothing to do with SO, 


If you have optimized your system in PBE and it "works", you don't need 
DFT-D3 (which is anyway often problematic for heavier atoms or TMs).


Am 20.06.2023 um 11:02 schrieb Burhan Ahmed:


Yes,

 1. I have optimized the bulk cell (with PBE) without so or mbj
 2. The slab is constructed from the optimized cell. I have generated
a supercell 1x1x2 with a vacuum of 40ang and from that a slab of
6ql (quintuple layers) is generated. The space group then changes
from 166-R3m to 164-p3m1 keeping the same hexagonal cell.
 3. After that I have minimized the internal atomic position
(relaxation) using msr1 method as available in wine2k code.

Now my question is that, I want to calculate the band structure for 
the slab with SO coupling. The steps I have mentioned are the correct 
one or I have to use the DFT-D3 correction. If DFT-D3 is a must one 
for the topological band structure then what modification/new steps 
required for the calculation.


Regards

Burhan Ahmed

*Research Scholar, AUS *

*From: *Laurence Marks 
*Sent: *Tuesday, June 20, 2023 9:48 AM
*To: *A Mailing list for WIEN2k users 

*Subject: *Re: [Wien] Calculation of surface state using lmbj 
potential with SO coupling.


A very basic question for you -- what is your surface structure? In 
general, just taking the bulk and cutting it is wrong (too lazy).


Before what you said:

a) Optimize the volume of the bulk cell with your functional of 
choice, no so or mbj.


b) Create a representative (e.g. valence neutral) surface slab using 
the 2D lattice parameters in the surface plane from a).


c) Optimise the internal positions (-min) as there will be some 
relaxation. Again no so or mbj. Check that you have, for instance, an 
insulator. You may have to recreate the surface if you did it wrong. 
Depending on case you will use run_lapw, runsp_c_lapw or runsp_lapw. 
Adding DW is optional.


Then start with adding mbj & so on. You have to keep your fingers 
crossed that the structure from c) is a valid representation, since 
you cannot optimize positions with so & mbj.


Depending upon care, 50% probability that you surface structure will 
be different from nature.


--
Professor Laurence Marks (Laurie)
Department of Materials Science and Engineering, Northwestern University
www.numis.northwestern.edu 
"Research is to see what everybody else has seen, and to think what 
nobody else has thought" Albert Szent-Györgyi


On Tue, Jun 20, 2023, 07:02 Burhan Ahmed  wrote:

Dear experts, I am having a compact cluster with 100gb RAM, 36
cores and 12 TB SATA hard disk with CentOS.

I want to calculate the band structure at the surface of
topological insulator. Below are the steps I have used for the
detailed calculation.

After initialization, I run a pbe scf using

“run_lapw -p -ec 0.0001 -cc 0.001 -NI”

Save the result using

“save_lapw *pbe*”

In the next step

“init_so_lapw” ( took the default parameters, magnetization is 0 0 1)

“Init_nl_vdw “ (kernel type 1)

“init_mbj_lapw”

“run_lapw -p -so -i 1 -NI”

“save_lapw *mbj*”

“init_mbj_lapw”

Took the selection 2 for surface calculation, and lmbj for potential

And finally I run “run_lapw -p -lmbj -so -ec 0.0001 -cc 0.001 -i
999 -NI”

And then I save the result using save_lapw *lmbj*.

I want to know whether these are the correct steps for calculating
band structure at the surface or I have to used DFT-D3 for the
correction of dispersion which are basically used for the surface
calculation. If DFT-D3 must be used for the surface calculation to
show Dirac Cone/band inversion nature, then what modification I
have to do in my steps to use DFT-D3 so that I can get a clear
Band structure with Dirac Cone.

Thanks in Advance

Regards

Burhan Ahmed

*Research Scholar, AUS *

___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at: 
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html



___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST 
at:http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html


--
---
Peter Blaha,  Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-158801165300
Email:peter.bl...@tuwien.ac.at   
WWW:http://www.imc.tuwien.ac.at   WIEN2k:http://www.wien2k.at

-

Re: [Wien] Calculation of surface state using lmbj potential with SO coupling.

[Laurence Marks]

Also, 40 Angers vacuum is killing defenceless electrons. I use 1/3 of the
cell, roughly 15-20 Angers.

---
Professor Laurence Marks (Laurie)
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu
"Research is to see what everybody else has seen, and to think what nobody
else has thought" Albert Szent-Györgyi

On Tue, Jun 20, 2023, 12:11 Laurence Marks  wrote:

> Do you know that your surface model is correct? Is it stoichiometric, a
> metal, and are the two surfaces the same or different? Is there a dipole?
>
> Remember GIGO.
>
> ---
> Professor Laurence Marks (Laurie)
> Department of Materials Science and Engineering
> Northwestern University
> www.numis.northwestern.edu
> "Research is to see what everybody else has seen, and to think what nobody
> else has thought" Albert Szent-Györgyi
>
> On Tue, Jun 20, 2023, 12:02 Burhan Ahmed  wrote:
>
>> Yes,
>>
>>1. I have optimized the bulk cell (with PBE) without so or mbj
>>2. The slab is constructed from the optimized cell. I have generated
>>a supercell 1x1x2 with a vacuum of 40ang and from that a slab of 6ql
>>(quintuple layers) is generated. The space group then changes from 166-R3m
>>to 164-p3m1 keeping the same hexagonal cell.
>>3. After that I have minimized the internal atomic position
>>(relaxation) using msr1 method as available in wine2k code.
>>
>>
>>
>> Now my question is that, I want to calculate the band structure for the
>> slab with SO coupling. The steps I have mentioned are the correct one or I
>> have to use the DFT-D3 correction. If DFT-D3 is a must one for the
>> topological band structure then what modification/new steps required for
>> the calculation.
>>
>>
>>
>> Regards
>>
>> Burhan Ahmed
>>
>> *Research Scholar, AUS *
>>
>>
>>
>>
>> Wien mailing list
>> Wien@zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>> SEARCH the MAILING-LIST at:
>> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>>
>
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] Calculation of surface state using lmbj potential with SO coupling.

[Laurence Marks]

Do you know that your surface model is correct? Is it stoichiometric, a
metal, and are the two surfaces the same or different? Is there a dipole?

Remember GIGO.

---
Professor Laurence Marks (Laurie)
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu
"Research is to see what everybody else has seen, and to think what nobody
else has thought" Albert Szent-Györgyi

On Tue, Jun 20, 2023, 12:02 Burhan Ahmed  wrote:

> Yes,
>
>1. I have optimized the bulk cell (with PBE) without so or mbj
>2. The slab is constructed from the optimized cell. I have generated a
>supercell 1x1x2 with a vacuum of 40ang and from that a slab of 6ql
>(quintuple layers) is generated. The space group then changes from 166-R3m
>to 164-p3m1 keeping the same hexagonal cell.
>3. After that I have minimized the internal atomic position
>(relaxation) using msr1 method as available in wine2k code.
>
>
>
> Now my question is that, I want to calculate the band structure for the
> slab with SO coupling. The steps I have mentioned are the correct one or I
> have to use the DFT-D3 correction. If DFT-D3 is a must one for the
> topological band structure then what modification/new steps required for
> the calculation.
>
>
>
> Regards
>
> Burhan Ahmed
>
> *Research Scholar, AUS *
>
>
>
>
> Wien mailing list
> Wien@zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] Calculation of surface state using lmbj potential with SO coupling.

[Burhan Ahmed]

Yes,I have optimized the bulk cell (with PBE) without so or mbj The slab is constructed from the optimized cell. I have generated a supercell 1x1x2 with a vacuum of 40ang and from that a slab of 6ql (quintuple layers) is generated. The space group then changes from 166-R3m to 164-p3m1 keeping the same hexagonal cell. After that I have minimized the internal atomic position (relaxation) using msr1 method as available in wine2k code.  Now my question is that, I want to calculate the band structure for the slab with SO coupling. The steps I have mentioned are the correct one or I have to use the DFT-D3 correction. If DFT-D3 is a must one for the topological band structure then what modification/new steps required for the calculation.   RegardsBurhan AhmedResearch Scholar, AUS  From: Laurence MarksSent: Tuesday, June 20, 2023 9:48 AMTo: A Mailing list for WIEN2k usersSubject: Re: [Wien] Calculation of surface state using lmbj potential with SO coupling. A very basic question for you -- what is your surface structure? In general, just taking the bulk and cutting it is wrong (too lazy). Before what you said:a) Optimize the volume of the bulk cell with your functional of choice, no so or mbj.b) Create a representative (e.g. valence neutral) surface slab using the 2D lattice parameters in the surface plane from a).c) Optimise the internal positions (-min) as there will be some relaxation. Again no so or mbj. Check that you have, for instance, an insulator. You may have to recreate the surface if you did it wrong. Depending on case you will use run_lapw, runsp_c_lapw or runsp_lapw. Adding DW is optional. Then start with adding mbj & so on. You have to keep your fingers crossed that the structure from c) is a valid representation, since you cannot optimize positions with so & mbj. Depending upon care, 50% probability that you surface structure will be different from nature. --Professor Laurence Marks (Laurie)Department of Materials Science and Engineering, Northwestern Universitywww.numis.northwestern.edu"Research is to see what everybody else has seen, and to think what nobody else has thought" Albert Szent-Györgyi On Tue, Jun 20, 2023, 07:02 Burhan Ahmed  wrote:Dear experts, I am having a compact cluster with 100gb RAM, 36 cores and 12 TB SATA hard disk with CentOS.I want to calculate the band structure at the surface of topological insulator. Below are the steps I have used for the detailed calculation.After initialization, I run a pbe scf using “run_lapw -p -ec 0.0001 -cc 0.001 -NI”Save the result using “save_lapw pbe”In the next step “init_so_lapw” ( took the default parameters, magnetization is 0 0 1)“Init_nl_vdw “ (kernel type 1)“init_mbj_lapw”“run_lapw -p -so -i 1 -NI”“save_lapw mbj”“init_mbj_lapw”Took the selection 2 for surface calculation, and lmbj for potential And finally I run “run_lapw -p -lmbj -so -ec 0.0001 -cc 0.001 -i 999 -NI”And then I save the result using save_lapw lmbj. I want to know whether these are the correct steps for calculating band structure at the surface or I have to used DFT-D3 for the correction of dispersion which are basically used for the surface calculation. If DFT-D3 must be used for the surface calculation to show Dirac Cone/band inversion nature, then what modification I have to do in my steps to use DFT-D3 so that I can get a clear Band structure with Dirac Cone. Thanks in Advance  RegardsBurhan AhmedResearch Scholar, AUS  ___Wien mailing listWien@zeus.theochem.tuwien.ac.athttp://zeus.theochem.tuwien.ac.at/mailman/listinfo/wienSEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html 
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] ** testerror: Error in Parallel LAPW

[Peter Blaha]

Well, the important files (output and error log files from slurm, the 
dayfile and all error files, .machines) are not present.


Also, in this way it is nearly impossible to see which files are recent 
and which are older ones.



The scripts are so complicated, that I cannot follow them  Why are 
you linking all wien-executables ?? You can set a PATH (in case the 
environment is not transferred).



Anyway, in the files I found some    lapw1.error_SAVED  file . It contains


'SELECT' - E-bottom -520.0 E-top 1.94197

'SELECT' - no energy limits found for atom 2 L= 0


I don't know if this is from the serial or parallel run, but in any 
case, this error has nothing to do with parallelization.


Either you used a different case.in1 file or  ?


The error you reported is a follow up error because lapw1 did not run 
properly, at least, this energy_1 file should be produced by lapw1.




Am 20.06.2023 um 10:18 schrieb Ilias Miroslav, doc. RNDr., PhD.:

Hello,

I am able to run serial SCF via SLURM
https://github.com/miroi/open-collection/blob/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg/virgo_slurm_wien2kgnupar_fromdstart.01

but when trying parallel
https://github.com/miroi/open-collection/blob/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg/virgo_slurm_wien2kgnupar_fromdstart.02 



I get lapw2.error

'LAPW2' - can't open unit: 30
'LAPW2' -    filename: LvO2onQg.energy_1
**  testerror: Error in Parallel LAPW2

The file "LvO2onQg.energy" is correct in serial mode.

Seems that LvO2onQg.energy_1 file is not produces in parallel run ?

All files are 
https://github.com/miroi/open-collection/tree/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg


Best,

Miro

___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST 
at:http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html


--
---
Peter Blaha,  Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-158801165300
Email:peter.bl...@tuwien.ac.at   
WWW:http://www.imc.tuwien.ac.at   WIEN2k:http://www.wien2k.at

-
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

[Wien] ** testerror: Error in Parallel LAPW

[Ilias Miroslav, doc. RNDr., PhD.]

Hello,

I am able to run serial SCF via SLURM
https://github.com/miroi/open-collection/blob/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg/virgo_slurm_wien2kgnupar_fromdstart.01

but when trying parallel
https://github.com/miroi/open-collection/blob/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg/virgo_slurm_wien2kgnupar_fromdstart.02

I get lapw2.error

'LAPW2' - can't open unit: 30
'LAPW2' -filename: LvO2onQg.energy_1
**  testerror: Error in Parallel LAPW2

The file "LvO2onQg.energy" is correct in serial mode.

Seems that LvO2onQg.energy_1 file is not produces in parallel run ?

All files are 
https://github.com/miroi/open-collection/tree/master/theoretical_chemistry/software/wien2k/runs/LvO2_on_small_quartz/wien2k/LvO2onQg

Best,

Miro
___
Wien mailing list
Wien@zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:  
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html

Re: [Wien] Requesting gentle clarification on OMP_NUM_THREADS, run_lapw, volume optimisation and wien2k update.

[Peter Blaha]

Usually, you would NEVER edit a cif file, unless you know very well the 
corresponding rules.


You have two possibilities to modify a structure:

I 've seen that the cif file is from VESTA. VESTA allows you to remove 
atoms, create larger cells, 


Then export it as cif file and import it into WIEN2k. Then run the 
symmetry programs nn, sgroup and symmetry.



Alternatively, convert the cif into a struct file and use the wien2k 
tools (x supercell) and the structeditor to remove atoms.


In any case, it is your responsibility to create a "reasonable" model. 
You cannot expect that I search the literature to find out the typical 
vacancy concentration of HfO2. But it should be clear: for a 3% vacancy 
concentration you need a supercell with about 30 oxygen atoms, .



Am 20.06.2023 um 09:18 schrieb Pranjal Nandi:


Dear Peter and Laurence,

Thank you for your valuable inputs. I have some followup queries.

 1. In the attached CIF file, I want to do DFT simulation on two
situations. One when there is excess oxygen and one in less
oxygen. However, I am confused as which part of the cif file I
should modify/implement the changes
 1. If I add a new line and add more oxygen, I am unsure of what I
should set in other parameters such as label, occupancy,
fract_x,y,z etc.
 2. I am also confused on how many additions/removal can be
considered as significant addition/reduction.
 2. When I am adding/subtracting oxygen, for realistic results
(matching simulation with lab experimental data), I should be
doing the simulations on a large supercell or unit cell (for
oxygen vacancies, I know that I should do a supercell as I cannot
create it in the unit cell cif file)?

Once again, thank you so much for your continuous support to the 
community.


With warm regards,

Pranjal

*From:*Wien  *On Behalf Of 
*Peter Blaha

*Sent:* Thursday, June 15, 2023 5:11 PM
*To:* wien@zeus.theochem.tuwien.ac.at
*Subject:* Re: [Wien] Requesting gentle clarification on 
OMP_NUM_THREADS, run_lapw, volume optimisation and wien2k update.


In addition to what L. Marks has already said:

A unit cell with 216 atoms should be run in mpi parallel mode. You 
won't have too many k-points probably, in particularly when doing 
position optimization.


About OMP Laurence Marks has already answered. The Intel mkl does not 
scale to higher OMP numbers. But: Test it out yourself. Set 
OMP_NUM_TREADS to 1 and run one iteration. Check the dayfile how long 
it takes. Set it to 2,4,8,16 and check the timing  You should see 
that lapw0 will benefit when you set it to 64 (for your 216 atom 
cell), but unfortunately not much speedup in lapw1/2 anymore.


Update of WIEN2k: You should install wien2k into a new directory, thus 
you do not destroy the old installation. In addition, there is a 
"copy" function in siteconfig which would copy the old settings to the 
new installation. It is briefly described on the download page .and in 
the (new !) usersguide as well as in the "update" site of wien2k.


Updating is strongly recommended since the new init_lapw has some 
automatic precision settings and you won't run your supercell with 
1000 k-points .


Volume optimization for high vacancy concentration is ok. What you 
read in the UG concerns the stress tensor, which is only NREL.


Hf-4f: No, these are core/semicore states. You don't need 
onsite-hybrid or LDA+U. However, depending on your RMTs you may be 
able to put them into the core by specifying a charge criterium 
instead of an energy (like   init -ecut 0.999)


I have some general queries and it would be great if you can
kindly provide me your feedback.




 1. I have 64 CPU (32 x 2) and more than 9 TB space and the system
is dedicated entirely to do simulations on DFT. I was thinking
to set the OMP_NUM_THREADS at 40 or 50 (even higher) to use
the maximum potential of the computer. But in the guidebook,
it is recommended that I do not set it higher than 8. May I
please the reason for this recommendation?


 2. Is there a code to run_lapw continuously till it reaches the
convergence and not start it again after 40 cycles?


 3. If I do a volume optimization for a single unit cell and then
I create a supercell of the optimized structure. Does the
supercell remains in optimized volume condition? Or it is
recommended that I do a volume optimization for the supercell
as well. Further is WIEN2k a reliable platform for volume
optimization of HfO as in the guidebook on page 87 it mentions
“The accuracy is thus limited to first and second row elements
(maybe early 3d elements) and in particular for the 5d series
and above, large errors may occur due to the neglect of
(scalar-) relativistic corrections.”


 4. Lastly, as it says on the website that it is highly
recommended to update the WIEN2K 

Re: [Wien] Requesting gentle clarification on OMP_NUM_THREADS, run_lapw, volume optimisation and wien2k update.

[Pranjal Nandi]

Dear Peter and Laurence,

Thank you for your valuable inputs. I have some followup queries.

  1.  In the attached CIF file, I want to do DFT simulation on two situations. 
One when there is excess oxygen and one in less oxygen. However, I am confused 
as which part of the cif file I should modify/implement the changes
 *   If I add a new line and add more oxygen, I am unsure of what I should 
set in other parameters such as label, occupancy, fract_x,y,z etc.
 *   I am also confused on how many additions/removal can be considered as 
significant addition/reduction.
  2.  When I am adding/subtracting oxygen, for realistic results (matching 
simulation with lab experimental data), I should be doing the simulations on a 
large supercell or unit cell (for oxygen vacancies, I know that I should do a 
supercell as I cannot create it in the unit cell cif file)?
Once again, thank you so much for your continuous support to the community.

With warm regards,
Pranjal

From: Wien  On Behalf Of Peter Blaha
Sent: Thursday, June 15, 2023 5:11 PM
To: wien@zeus.theochem.tuwien.ac.at
Subject: Re: [Wien] Requesting gentle clarification on OMP_NUM_THREADS, 
run_lapw, volume optimisation and wien2k update.


In addition to what L. Marks has already said:

A unit cell with 216 atoms should be run in mpi parallel mode. You won't have 
too many k-points probably, in particularly when doing position optimization.

About OMP Laurence Marks has already answered. The Intel mkl does not scale to 
higher OMP numbers. But: Test it out yourself. Set OMP_NUM_TREADS to 1 and run 
one iteration. Check the dayfile how long it takes. Set it to 2,4,8,16 and 
check the timing  You should see that lapw0 will benefit when you set it to 
64 (for your 216 atom cell), but unfortunately not much speedup in lapw1/2 
anymore.

Update of WIEN2k: You should install wien2k into a new directory, thus you do 
not destroy the old installation. In addition, there is a "copy" function in 
siteconfig which would copy the old settings to the new installation. It is 
briefly described on the download page .and in the (new !) usersguide as well 
as in the "update" site of wien2k.

Updating is strongly recommended since the new init_lapw has some automatic 
precision settings and you won't run your supercell with 1000 k-points .

Volume optimization for high vacancy concentration is ok. What you read in the 
UG concerns the stress tensor, which is only NREL.

Hf-4f: No, these are core/semicore states. You don't need onsite-hybrid or 
LDA+U. However, depending on your RMTs you may be able to put them into the 
core by specifying a charge criterium instead of an energy (like   init -ecut 
0.999)



I have some general queries and it would be great if you can kindly provide me 
your feedback.




  1.  I have 64 CPU (32 x 2) and more than 9 TB space and the system is 
dedicated entirely to do simulations on DFT. I was thinking to set the 
OMP_NUM_THREADS at 40 or 50 (even higher) to use the maximum potential of the 
computer. But in the guidebook, it is recommended that I do not set it higher 
than 8. May I please the reason for this recommendation?


  2.  Is there a code to run_lapw continuously till it reaches the convergence 
and not start it again after 40 cycles?


  3.  If I do a volume optimization for a single unit cell and then I create a 
supercell of the optimized structure. Does the supercell remains in optimized 
volume condition? Or it is recommended that I do a volume optimization for the 
supercell as well. Further is WIEN2k a reliable platform for volume 
optimization of HfO as in the guidebook on page 87 it mentions “The accuracy is 
thus limited to first and second row elements (maybe early 3d elements) and in 
particular for the 5d series and above, large errors may occur due to the 
neglect of (scalar-) relativistic corrections.”


  4.  Lastly, as it says on the website that it is highly recommended to update 
the WIEN2K version, I am unable to find the manual where I can follow the 
instructions to update the version. As I have not done any installation of 
WIEN2K (my past colleague installed the present version), I am worried that 
while upgrading I may commit some mistake. A guideline on updating would have 
been very helpful.


Looking forward to your kind feedback.

Thank you in advance.

With warm regards,
Pranjal




Aquest missatge, i els fitxers adjunts que hi pugui haver, pot contenir 
informació confidencial o protegida legalment i s’adreça exclusivament a la 
persona o entitat destinatària. Si no consteu com a destinatari final o no 
teniu l’encàrrec de rebre’l, no esteu autoritzat a llegir-lo, retenir-lo, 
modificar-lo, distribuir-lo, copiar-lo ni a revelar-ne el contingut. Si l’heu 
rebut per error, informeu-ne el remitent i elimineu del sistema tant el 
missatge com els fitxers adjunts que hi pugui haver.

Este mensaje, y los ficheros adjuntos que pueda incluir, puede contener 
información confidencial o