[Wien] Restart

[Saeid Jalali]

1. Save your last 20 performed SCF cycles to keep case.clmsum, case.scf, case.dmat, case.vorb and case.struct under your desired filename, using save command that also removes case.broyd* files: save_lapw filename
  2. Rerun after your last saved calculations using run command: run_lapw -i 90 -cc 0.00001


  Here -i 90 indicates to be increased the ubmer of max. iterations from the default value of 20 to 90.
  The -s switch is not needed here as by default run_lapw always starts from the first point, i.e. lapw0. Actually the -s switch is useful for the case that one intends to start the program from another point than lapw0. The -NI switch does not also matter after the save_lapw command, becuse as it was stated in step 1 the save command removes case.broyd* files. Thus -NI cannot keep the removed files!
  
Manish Singh <mksingh at gmail.com> wrote:
  Dear Wien users,

I know there was a thread on this one long time ago (please see below). I
seek a clarification regarding restarting a job. I have a supercell case
(complex calculation) which stopped after 20 out of 20 iterations. There
were no errors but since the convergence had not been achieved (as specified
in the dayfile, please see below), I would have to restart the job.
The last iteration was completed.
Now, how do I restart the calculation from the point it stopped:

Should I use 

run_lapw -s lapw0

because the next iteration would start from lapw0?

Should I use the -NI flag too (so as to continue from the same charge
history)? Or would the -s option take care of it on its own?

Also, would the -s option take care of the fact that the calculaton is
complex (lapw1 & lapw2)?

Regards,
Manish


cycle 20 (Fri Jul 14 05:00:58 CDT 2006) (1/1 to go)

> lapw0 (05:00:58) 0.2u 0.1s 0:15 2% 355+937k 0+0io 8pf+0w
> lapw1 -c (05:01:13) 0.2u 0.1s 0:15 2% 360+928k 0+0io 7pf+0w
> lapw2 -c (05:01:29) 0.2u 0.1s 0:15 2% 364+882k 0+0io 0pf+0w
> lcore (05:01:44) 0.2u 0.0s 0:15 2% 361+917k 0+0io 0pf+0w
> mixer (05:02:08) 0.2u 0.1s 0:10 3% 352+878k 0+0io 0pf+0w
:ENERGY convergence: 0 0 0
:CHARGE convergence: 0 0.00001 .0001390
restart
ec cc and fc_conv 1 0 1

> stop


-----Original Message-----
From: Peter Blaha [mailto:wien at zeus.theochem.tuwien.ac.at] 
Sent: Wednesday, August 13, 2003 12:27 PM
Subject: [Wien] Restarting the calculation

> I have a large case with spin-orbit coupling. It will take several days to
run a single lapw1 step in SCF calculation. After one lapw1 step finishes, I
met a limited memory problem (i.e. the assigned memory is less than the one
during the case running ) when lapwso is running. So the program stoped due
to the memory problem. My question is how can I restart the SCF right where
it stoped after I increase the memory? Is this possible?
>
run_lapw -so -s lapwso

You can start at any "label" in run_lapw.
Have a look into this script. A "label" is

word:


P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671 FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at WWW:
http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------


_______________________________________________
Wien mailing list
Wien at zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien



Sincerely yours,
S. Jalali.
/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
Saeid Jalali Asadabadi,
Department of Physics, Faculty of Science,
University of Isfahan (UI), Hezar Gerib Avenue,
81744 Isfahan, Iran.
Phones:
Dep. of Phys.      :+98-0311-793 2435
Office             :+98-0311-793 2430
Fax No.            :+98-0311-793 2409
E-mail             :s_jalali_a at yahoo.com
www                :http://www.ui.ac.ir
/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
 		
---------------------------------
How low will we go? Check out Yahoo! Messenger’s low  PC-to-Phone call rates.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20060714/f0f98f34/attachment.html