[gpaw-users] calculation with UnitCellFilter

[Marcin Dulak]

Hi,

On 12/07/12 11:46, Vladislav Ivanistsev wrote:
> Dear Marcin,
> you might be right.
> Recently I've started to mention that a lot of calculations won't 
> converge or run properly on available supercomputer using 12CPU per 
> core. Is it somehow systematic that GPAW requires 8x cores to run a job?
this should not be the case. Can you run one of your failing jobs on 8 
cores?

Marcin

> Thank you very much
> Vladilsav
>
>
> On Fri, Dec 7, 2012 at 9:19 AM, Marcin Dulak 
> <Marcin.Dulak at fysik.dtu.dk <mailto:Marcin.Dulak at fysik.dtu.dk>> wrote:
>
>     Hi,
>
>
>     On 12/05/12 15:33, Vladislav Ivanistsev wrote:
>>     Dear Marcin,
>>
>>     my input file and output are attached.
>>     The small t.py example worked.
>     your example works on our cluster (4 cores,  16 cores).
>     Could it be something external terminating your job?
>     Have you run all parallel tests on this machine?
>     mpiexec -np 8 gpaw-python `which gpaw-test` 2>&1 | tee test.log
>
>     Marcin
>
>
>>     Thankfully,
>>     Vladislav Ivanistsev
>>
>>
>>     On Wed, Dec 5, 2012 at 11:27 AM, Vladislav Ivanistsev
>>     <olunet at gmail.com <mailto:olunet at gmail.com>> wrote:
>>
>>         Hi,
>>         the small example works. In my crystal the internal structure
>>         must be also optimized. Might it be the reason?
>>
>>
>>
>>         On Wed, Dec 5, 2012 at 9:19 AM, Marcin Dulak
>>         <Marcin.Dulak at fysik.dtu.dk
>>         <mailto:Marcin.Dulak at fysik.dtu.dk>> wrote:
>>
>>             Hi,
>>
>>             does the small example (t.py) i attached in my first
>>             answer work?
>>
>>             Marcin
>>
>>
>>             On 12/04/12 23:08, Vladislav Ivanistsev wrote:
>>>             Dear Marcin,
>>>             with and without saving to traj, the calculation ends
>>>             with the same notification:
>>>             mpirun noticed that process rank 2 with PID 24466 on
>>>             node mts-Precision-T5600 exited on signal 9 (Killed).
>>>             No other errors. Calculation converged. I was using 4
>>>             and 24 CPU for testing.
>>>             Could you please look for the possible source of the error.
>>>             Many thanks,
>>>             Vladislav ivanistsev
>>>
>>>
>>>             On Mon, Dec 3, 2012 at 5:53 PM, Marcin Dulak
>>>             <Marcin.Dulak at fysik.dtu.dk
>>>             <mailto:Marcin.Dulak at fysik.dtu.dk>> wrote:
>>>
>>>                 On 12/03/12 15:43, Ask Hjorth Larsen wrote:
>>>                 > Hi
>>>                 >
>>>                 > 2012/12/3 Marcin Dulak <Marcin.Dulak at fysik.dtu.dk
>>>                 <mailto:Marcin.Dulak at fysik.dtu.dk>>:
>>>                 >> Hi,
>>>                 >>
>>>                 >> the problem is one cannot use trajectory directly
>>>                 with UnitCellFilter (and
>>>                 >> similar),
>>>                 >> as with "standard" positions optimizer. See a
>>>                 workaround example attached
>>>                 >>
>>>                 (https://wiki.fysik.dtu.dk/ase/tutorials/lattice_constant.html#using-the-stress-tensor).
>>>                 >>
>>>                 >> Anybody knows how to implement an error message
>>>                 in such cases?
>>>                 i mean the cases which do not support trajectory
>>>                 keyword when creating
>>>                 an optimizer instance.
>>>
>>>                 Marcin
>>>
>>>
>>>                 >>
>>>                 >> Marcin
>>>                 > Which cases?
>>>                 >
>>>                 > Regards
>>>                 > Ask
>>>                 >
>>>                 >
>>>
>>>
>>>                 --
>>>                 ***********************************
>>>
>>>                 Marcin Dulak
>>>                 Technical University of Denmark
>>>                 Department of Physics
>>>                 Building 307, Room 229
>>>                 DK-2800 Kongens Lyngby
>>>                 Denmark
>>>                 Tel.: (+45) 4525 3157
>>>                 Fax.: (+45) 4593 2399
>>>                 email: Marcin.Dulak at fysik.dtu.dk
>>>                 <mailto:Marcin.Dulak at fysik.dtu.dk>
>>>
>>>                 ***********************************
>>>
>>>                 _______________________________________________
>>>                 gpaw-users mailing list
>>>                 gpaw-users at listserv.fysik.dtu.dk
>>>                 <mailto:gpaw-users at listserv.fysik.dtu.dk>
>>>                 https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>>>
>>>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listserv.fysik.dtu.dk/pipermail/gpaw-users/attachments/20121207/210e8c71/attachment.html