[Wien] Insufficient virtual memory - LAPW2

[Laurence Marks]

Do "grep -e RKM *.scf* " which will give you the size of the matrix.
Also, how many k-points?  I suspect that both is are large in which
case it is certainly possible to request too much memory for a 8Gb
machine although I am surprised that it fails with a 64Gb machine. If
RKM is large you may need to move to the mpi version and use
lapw2_vector_split:2 .

If it is not this, then there is something wrong at the OS level
and/or some other code is running. One way to look is to use top while
lapw2 is running (e.g. do x lapw2 from a terminal, and top in another
terminal).

N.B., the default number of k-points is a reasonable estimate for a
small unit cell metal. In general the number of k-point needed is
going to scale as 1/Cell Volume, i.e. one needs a certain density of
k-points in reciprocal space.


On Tue, Oct 15, 2013 at 1:53 AM, Lyudmila Dobysheva <lyuka17 at mail.ru> wrote:
> 14.10.2013 18:33, Salman Zarrini wrote:
>
> Dear Salman,
>
> I think your case is very strange, so, being into despair, till our
> experts think, let's precise:
> 1)
>  > there are no unlimited restriction
>  > on the machines. multiple (a, b and c)
>  > machines which have the same installation of WIEN2K-12 (OS and
>  > software)
>
> Fortran and mkl also are the same?
>
> 2)
>> a) 2 Socket, Dual Core AMD Opteron(tm) Processor 270, 8GB RAM
>> b) 2 Socket, (8Core) AMD Opteron(tm) Processor 6320, 64GB RAM
>> c) 2 Socket, (8Core(*2=HT)) Intel(R) Xeon(R) CPU E5-2680 0 @ 2.70GHz,
>> 256GB RAM
>> $ ulimit -a
>> ....
>
> I imply here that you have done the same for all three computers and
> obtained the same answers (?)
>
> 3)
> In some configurations, as far as I remember, the system does not
> transfer set-up to executing scripts. I'd advice you to add the command
> "ulimit -a" to run_lapw (or runsp_lapw if you have used this) and to
> check the output for each computer, no necessity to really make the
> cycle, it is sufficient to add this command somewhere in the beginning
> around lapw0 execution and a stop command just after it, and to see the
> output.
>
> 4)
>> on machines (a) and (b) failed
>> (a) failed in the first iteration
>> (b) failed in the third iteration
> This (b) computer completes two iterations. Compare scf file with the
> scf file from (c) computer. Are these two iterations identical?
>
> 5)
> During these two iterations make "top" in terminal and check what
> happened with memory during lapw2. And maybe you see something during
> lapw2 in the third iteration.
>
> For the present, nothing more clever comes into my mind.
>
> Best wishes
>    Lyudmila Dobysheva
> ------------------------------------------------------------------
> Phys.-Techn. Institute of Ural Br. of Russian Ac. of Sci.
> 426001 Izhevsk, ul.Kirova 132
> RUSSIA
> ------------------------------------------------------------------
> Tel.:7(3412) 442118 (home), 218988(office), 722529(Fax)
> E-mail: lyu at ftiudm.ru
>          lyuka17 at mail.ru (office) lyuka17 at gmail.com (home)
> Skype:  lyuka17 (home), lyuka18 (office)
> http://fti.udm.ru/content/view/25/103/lang,english/
> ------------------------------------------------------------------
> _______________________________________________
> Wien mailing list
> Wien at 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



-- 
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu 1-847-491-3996
"Research is to see what everybody else has seen, and to think what
nobody else has thought"
Albert Szent-Gyorgi