[Nek5000-users] Maximum number of history points using hpts()

nek5000-users at lists.mcs.anl.gov nek5000-users at lists.mcs.anl.gov
Fri Mar 30 04:40:48 CDT 2018


A Nek5000 run on NP ranks with (in total) 359640 history points requires lhis >= 359640/NP + 1.
I don't see why with lhis = 359641 you need more than 1 MPI ranks. At least from a history point memory perspective. 

-----Original message-----
> From:nek5000-users at lists.mcs.anl.gov <nek5000-users at lists.mcs.anl.gov>
> Sent: Friday 30th March 2018 10:30
> To: nek5000-users at lists.mcs.anl.gov
> Subject: Re: [Nek5000-users] Maximum number of history points using hpts()
> 
>     I have increased the number of MPI ranks and it worked just fine.
>     Thank you,
>     
>     Best regards,
>     
>     Arnold
>     
>     
> Le 30/03/2018 à 09:48,
>       nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> a écrit :
> Hi so lhis is equal to 359641 and I'm using 7 ranks.. However I
>         can try and increase that value (which I will try immediately).
>         
>         Best,
> Arnold
>       
> Le 29/03/2018 à 17:57, nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> a
>         écrit :
> What's your lhis? How many MPI ranks are you using?
> -----Original message-----
> From:nek5000-users at lists.mcs.anl.gov <mailto:From:nek5000-users at lists.mcs.anl.gov> <nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>Sent: Thursday 29th March 2018 17:54
> To: nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov>Subject: Re: [Nek5000-users] Maximum number of history points using hpts()
> I'm executing 9 times nek5000 to call hpts (using different
>       case.his files containing each 39960 points, using a shell script)
>       because I need When I'm using 359640=9*39660 points 
> This works perfectly fine. However when I'm trying to use a
>       single file in order to execute nek5000 once, it gives the
>       following error following error when the number of points is equal
>       to 359640 : 
>       dump history points
>        reading history points
>        found       359640  points
>       APPLICATION TERMINATED WITH THE EXIT STRING: Segmentation fault
>       (signal 11)
> I guess I'm encountering some memory problems..
> Best regards,
> Arnold
> Le 29/03/2018 à 17:27,
>       nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> a écrit :
> What do you mean with doesn't work anymore. How does it manifest itself?
> -----Original message-----
> From:nek5000-users at lists.mcs.anl.gov <mailto:From:nek5000-users at lists.mcs.anl.gov> <mailto:From:nek5000-users at lists.mcs.anl.gov> <mailto:From:nek5000-users at lists.mcs.anl.gov> <nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>Sent: Thursday 29th March 2018 17:23
> To: nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov>Subject: Re: [Nek5000-users] Maximum number of history points using hpts()
> Hello Pierre, 
> In my case my case.his file contains 39960 points and lhis=39961,
>       however when I increased the first value and the lhis variable
>       accordingly it doesn't work anymore. Is there something I'm
>       missing?
> Best,
> Arnold
> Le 29/03/2018 à 16:30,
>       nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> a écrit :
>     Hi
>       Arnold,
>       You can set it in the SIZE file:
>       param(lhis=100) ! # of pts a proc reads from hpts.in
>                                   ! Note: lhis*np > npoints in
>       hpts.in
>       best,
>       Pierre
>       On 2018-03-29 16:14, nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> <mailto:nek5000-users at lists.mcs.anl.gov> wrote:
>       Hi Neks,
>         I was wondering if there is any maximum number of history points
>         one should put in the case.his file. If someone could help me
>         regarding that topic, I haven't found any information yet.
>         Thank you for your time,
>         Best,
>         Arnold
>         _______________________________________________
>         Nek5000-users mailing list
>         Nek5000-users at lists.mcs.anl.gov <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>        https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>      _______________________________________________
>       Nek5000-users mailing list
>       Nek5000-users at lists.mcs.anl.gov <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>      https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>-- 
> Arnold Wakim
>           Ph.D. Student
>           Département Aérodynamique, Aéroélasticité, Acoustique
>           INST
>           Tél: +33 1 46 23 51 83
>           ONERA - The French Aerospace Lab - Centre de Meudon
>           8, rue des Vertugadins - 92190 MEUDON
>           Nous suivre sur : www.onera.fr <http://www.onera.fr> <http://www.onera.fr> <http://www.onera.fr> <http://www.onera.fr> <http://www.onera.fr> <http://www.onera.fr> <http://www.onera.fr> | Twitter <http://www.twitter.com/@onera_fr> <http://www.twitter.com/@onera_fr> <http://www.twitter.com/@onera_fr> <http://www.twitter.com/@onera_fr> |  LinkedIn <http://www.linkedin.com/company/onera> <http://www.linkedin.com/company/onera> <http://www.linkedin.com/company/onera> <http://www.linkedin.com/company/onera> | Facebook <http://www.facebook.fr/thefrenchaerospacelab> <http://www.facebook.fr/thefrenchaerospacelab> <http://www.facebook.fr/thefrenchaerospacelab> <http://www.facebook.fr/thefrenchaerospacelab>          Avertissement/disclaimer http://www.onera.fr/onera-en/emails-terms <http://www.onera.fr/onera-en/emails-terms> <http://www.onera.fr/onera-en/emails-terms> <http://www.onera.fr/onera-en/emails-terms> <http://www.onera.fr/onera-en
 /emails-
 terms> <http://www.onera.fr/onera-en/emails-terms> <http://www.onera.fr/onera-en/emails-terms> <http://www.onera.fr/onera-en/emails-terms>_______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>_______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov> <mailto:Nek5000-users at lists.mcs.anl.gov>https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users> <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>-- 
> Arnold Wakim
>           Ph.D. Student
>           Département Aérodynamique, Aéroélasticité, Acoustique
>           INST
>           Tél: +33 1 46 23 51 83
>           ONERA - The French Aerospace Lab - Centre de Meudon
>           8, rue des Vertugadins - 92190 MEUDON
>           Nous suivre sur : www.onera.fr <http://www.onera.fr> <http://www.onera.fr> <http://www.onera.fr> | Twitter <http://www.twitter.com/@onera_fr> <http://www.twitter.com/@onera_fr> |  LinkedIn <http://www.linkedin.com/company/onera> <http://www.linkedin.com/company/onera> | Facebook <http://www.facebook.fr/thefrenchaerospacelab> <http://www.facebook.fr/thefrenchaerospacelab>          Avertissement/disclaimer http://www.onera.fr/onera-en/emails-terms <http://www.onera.fr/onera-en/emails-terms> <http://www.onera.fr/onera-en/emails-terms> <http://www.onera.fr/onera-en/emails-terms>_______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov <mailto:Nek5000-users at lists.mcs.anl.gov>https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>_______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov <mailto:Nek5000-users at lists.mcs.anl.gov>https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>      
> -- 
> Arnold Wakim
>             Ph.D. Student
>             Département Aérodynamique, Aéroélasticité, Acoustique
>             INST
>             Tél: +33 1 46 23 51 83
> 
>             ONERA - The French Aerospace Lab - Centre de Meudon
>             8, rue des Vertugadins - 92190 MEUDON
>             
>             Nous suivre sur : www.onera.fr <http://www.onera.fr> | Twitter <http://www.twitter.com/@onera_fr> | 
>               LinkedIn <http://www.linkedin.com/company/onera> | Facebook <http://www.facebook.fr/thefrenchaerospacelab>
>             Avertissement/disclaimer http://www.onera.fr/onera-en/emails-terms <http://www.onera.fr/onera-en/emails-terms>      
>       
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov <mailto:Nek5000-users at lists.mcs.anl.gov>https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users>    
> -- 
> Arnold Wakim
>           Ph.D. Student
>           Département Aérodynamique, Aéroélasticité, Acoustique
>           INST
>           Tél: +33 1 46 23 51 83
> 
>           ONERA - The French Aerospace Lab - Centre de Meudon
>           8, rue des Vertugadins - 92190 MEUDON
>           
>           Nous suivre sur : www.onera.fr <http://www.onera.fr> | Twitter <http://www.twitter.com/@onera_fr> |  LinkedIn <http://www.linkedin.com/company/onera> | Facebook <http://www.facebook.fr/thefrenchaerospacelab>
>           Avertissement/disclaimer http://www.onera.fr/onera-en/emails-terms <http://www.onera.fr/onera-en/emails-terms>_______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users


More information about the Nek5000-users mailing list