[Nek5000-users] Simulation hanging at 'mapping elements to processors'

nek5000-users at lists.mcs.anl.gov nek5000-users at lists.mcs.anl.gov
Sat Sep 23 00:21:17 CDT 2017 

You can use tools/amg_hypre instead of amg_matlab2. 

On 23 Sep 2017, at 02:15, "nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> " <nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> > wrote:

Hi,

the job run until it threw the following error message:

EXIT: Problem size requires AMG solver          1

I attached the log file.

My mesh has 478946 elements. Do I have to use the AMG solver, or is there a workaround? If I remember correctly the AMG solver requires Matlab which I do not have access to.

Thanks,
Marco


_______________________________________________

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



<Actesys.1.000>

On Sep 22, 2017, at 8:26 AM, nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> wrote:

I generated the ma2 file and submitted the job. I will let you know how it goes.

How different is the ma2 file from a map file?

Thanks,
Marco

On Sep 22, 2017, at 8:03 AM, nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> wrote:

This looks like a bug. 
Can you try with the latest version and run genmap again (this will produce a .ma2 file).  Make sure you have just a .ma2 and no .map in your case directory.

If this works I have to check why the old .map doesn’t.

Stefan

On 22 Sep 2017, at 13:56, "nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> " <nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> > wrote:

Stefan,

below is the log file of the run:

|      _   __ ______ __ __  ______  ____   ____   ____     |
|     / | / // ____// //_/ / ____/ / __ \\/ __ \\/ __ \\   |
|    /  |/ // __/  / ,<   /___ \\ / / / // / / // / / /    |
|   / /|  // /___ / /| | ____/ / / /_/ // /_/ // /_/ /     |
|  /_/ |_//_____//_/ |_|/_____/  \\___/ \\___/ \\___/      |
|                                                          |
|----------------------------------------------------------|
|                                                          |
| NEK5000:  Open Source Spectral Element Solver            |
| COPYRIGHT (c) 2008-2017 UCHICAGO ARGONNE, LLC            |
| Version:  17.0.0-dev                                     |
| Web:      http://nek5000.mcs.anl.gov <http://nek5000.mcs.anl.gov/>                      |
|                                                          |
\\----------------------------------------------------------/


Number of processors:        1024
REAL    wdsize      :           8
INTEGER wdsize      :           4
Timer accuracy      : 9.54E-08

Reading /home/mxd/Nek5000/run/actasys/3d-model/actuatorA.rea
Switch on stress formulation to support PN/PN and IFUSERVP=T
Reading /home/mxd/Nek5000/run/actasys/3d-model/actuatorA.re2
mapping elements to processors
Reading /home/mxd/Nek5000/run/actasys/3d-model/actuatorA.map

I run my case with 1024 processors as suggested by Paul. The run hung for two hours before reaching the wall time.

Marco
> On Sep 22, 2017, at 3:09 AM, nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> wrote:
> 
> Marco,
> 
> can you please post your logfile. What version of Nek5000 do you use?
> 
> Cheers,
> Stefan
> 
> -----Original message-----
>> From:nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> <nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> >
>> Sent: Friday 22nd September 2017 2:40
>> To: nek5000-users at lists.mcs.anl.gov <mailto:nek5000-users at lists.mcs.anl.gov> 
>> Subject: Re: [Nek5000-users] Simulation hanging at 'mapping elements to processors'
>> 
>> Hi Paul,
>> <br class="" />
>> thanks for the help. I will give it a try tomorrow. I am indeed using the XXT coarse grid solver.
>> <br class="" />
>> Marco
>> <br class="" />
>> On Sep 21, 2017, at 6:49 PM, 
>> 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> > <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:
>> <br class="Apple-interchange-newline" />
>> <br class="Apple-interchange-newline" />
>> Hi Marco,
>> <br class="" />
>> I'm not 100 % about the hanging point, but I suggest - especially for larger problems like this - running on a power of 2 processors (i.e., 1024 in your case).
>> <br class="" />
>> This is particularly critical for performance when the XXT (default) coarse grid solver is used.  Less so, perhaps, for the AMG coarse solver...I'm assuming you're using the XXT one - but not certain.
>> <br class="" />
>> Best,
>> <br class="" />
>> Paul
>> <br class="" />
>> -----------
>> From: Nek5000-users <nek5000-users-bounces at lists.mcs.anl.gov <mailto:nek5000-users-bounces at lists.mcs.anl.gov> <mailto:nek5000-users-bounces at lists.mcs.anl.gov <mailto:nek5000-users-bounces at lists.mcs.anl.gov> >>
>> on behalf of 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> > <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> >><br class="" />
>> Sent: Thursday, September 21, 2017 3:37:55 PM<br class="" />
>> 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> ><br class="" />
>> Subject: [Nek5000-users] Simulation hanging at 'mapping elements to processors'
>> Hello,<br class="" />
>> <br class="" />
>> I am trying to run a simulation that has ~500,000 elements on 1280 processors with lx1=4 for debugging purposes. My simulation hangs after outputting 'mapping elements to processors’. I generated the mesh with Cubit and then used the exo2nek script to get the
>> re2 file. Genmap runs fine on the re2 file.<br class="" />
>> <br class="" />
>> I was wondering if this problem has been encountered in the past, and what I could do to fix it.<br class="" />
>> <br class="" />
>> Thanks,<br class="" />
>> Marco<br class="" />
>> <br class="" />
>> _______________________________________________<br class="" />
>> Nek5000-users mailing list<br class="" />
>> 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> ><br class="" />
>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users><br class="" />
>> _______________________________________________<br style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;" class="" />
>> Nek5000-users
>> mailing list<br style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;" class="" />
>> 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> ><br style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;" class="" />
>> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users <https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users><br class="" />
>> _______________________________________________
>> 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
> _______________________________________________
> 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

_______________________________________________
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

_______________________________________________
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

_______________________________________________
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

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20170923/22662121/attachment-0001.html>

More information about the Nek5000-users mailing list