Hi Azad, <br><br>This error means that your version of matlab doesn't like the fprint() in coarse_par.f, line 37. <br><br>Since this is just a print statement, you can comment this out by adding % in front of line 37. So it should look like:<br>
<br>% fprintf(1,'%g\n',max(lanczos(sparse(D*S*D))));<br>
<br>Then this should work, at least for turbChannel. <br><br>good luck,<br>Katie<br><br><div class="gmail_quote">On Thu, Jan 12, 2012 at 2:34 PM, <span dir="ltr"><<a href="mailto:nek5000-users@lists.mcs.anl.gov">nek5000-users@lists.mcs.anl.gov</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><br>
Hi Katie;<br>
<br>
Thanks for updating the page. I did exactly the same as it's suggested in the updated page. After ./run of the script in amg_matlab I got this error:<br>
<br>
##############################<u></u>###########################<br>
Computing Lagrange multiplier governing matrix skeleton ... nnz = 25.<br>
Computing Lagrange multiplier governing matrix (C code) ... done.<br>
CG to obtain Lagrange multipliers ... 1 iterations.<br>
Computing interpolation weights (C code) ... done.<br>
Sparsifying R_f A P: compression = 0<br>
simple_sparsify: nnzs 4/4 (1)<br>
Level 6, dim(A) = 2<br>
Computing S = I - D^{-1/2} A D^{-1/2} ... done, nnz(S) = 2.<br>
Running coarsening heuristic ...<br>
ratio = 0, n = 0, max Gershgorin radius = 1<br>
ratio = 0.5, n = 1, max Gershgorin radius = 0<br>
connectivity = ??? Error using ==> fprintf<br>
Function is not defined for sparse inputs.<br>
<br>
Error in ==> coarse_par at 37<br>
fprintf(1,'%g\n',max(lanczos(sparse(D*S*D))));<br>
<br>
Error in ==> amg_setup at 30<br>
[C F] = coarse_par(A,tolc);<br>
<br>
Error in ==> go at 20<br>
data = amg_setup(A, full(0*A(:,1)+1),ctol, tol, stol, wtol);<br>
##############################<u></u>##############################<br>
<br>
I tried it for turbChannel test case as well as some other simplified test cases. I have got the same errors.<br>
<br>
<br>
Regards<br>
Azad<br>
<br>
<br>
<br>
Quoting <a href="mailto:nek5000-users-request@lists.mcs.anl.gov" target="_blank">nek5000-users-request@lists.<u></u>mcs.anl.gov</a>:<br>
<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Send Nek5000-users mailing list submissions to<br>
<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><br>
or, via email, send a message with subject or body 'help' to<br>
<a href="mailto:nek5000-users-request@lists.mcs.anl.gov" target="_blank">nek5000-users-request@lists.<u></u>mcs.anl.gov</a><br>
<br>
You can reach the person managing the list at<br>
<a href="mailto:nek5000-users-owner@lists.mcs.anl.gov" target="_blank">nek5000-users-owner@lists.mcs.<u></u>anl.gov</a><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of Nek5000-users digest..."<br>
<br>
<br>
Today's Topics:<br>
<br>
1. Re: run-time hang up in gs_setup (<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>)<br>
<br>
<br>
------------------------------<u></u>------------------------------<u></u>----------<br>
<br>
Message: 1<br>
Date: Thu, 12 Jan 2012 14:09:12 -0600<br>
From: <a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a><br>
Subject: Re: [Nek5000-users] run-time hang up in gs_setup<br>
To: <a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a><br>
Message-ID:<br>
<CAFTrYW+<u></u>ctDhLYVdJCh8mOpA6bB7j361o=<a href="mailto:vn7WaATYkEDCifYsw@mail.gmail.com" target="_blank">vn7W<u></u>aATYkEDCifYsw@mail.gmail.com</a>><br>
Content-Type: text/plain; charset="iso-8859-1"<br>
<br>
Hi Azad,<br>
<br>
Earlier you had asked if a step was missing in the wiki with regards to the<br>
amg files/scripts. The has just been updated.<br>
<br>
Can you tell me exactly what you did to run the turbChannel test? And what<br>
output you had from the amg_matlab/run step?<br>
<br>
thanks,<br>
Katie<br>
<br>
On Thu, Jan 12, 2012 at 1:38 PM, <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>> wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Hi Stefan;<br>
<br>
Unfortunately it does not work. first of all, ./run in amg_matlab does not<br>
produce anything (latest version). Using my version of the files (running<br>
turbChannel), the simulation crashed giving me this error:<br>
<br>
##############################<u></u>**#################<br>
AMG level 8: 3 iterations with rho = 0.680429<br>
AMG level 9: 2 iterations with rho = 0.560188<br>
AMG: 144540 rows<br>
AMG: reading through row 1200, pass 1/121<br>
ERROR (proc 0000, /scratch/azad/codes/late/nek5_<u></u>**svn/trunk/nek/jl/amg.c:875):<br>
AMG: data<br>
has more rows than given problem<br>
<br>
call exitt: dying ...<br>
##############################<u></u>**#################<br>
<br>
Regards<br>
Azad<br>
<br>
Quoting nek5000-users-request@lists.**<a href="http://mcs.anl.gov" target="_blank"><u></u>mcs.anl.gov</a><<a href="mailto:nek5000-users-request@lists.mcs.anl.gov" target="_blank">nek5000-users-<u></u>request@lists.mcs.anl.gov</a>><br>
:<br>
<br>
Send Nek5000-users mailing list submissions to<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
nek5000-users@lists.mcs.anl.**<u></u>gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
<a href="https://lists.mcs.anl.gov/**mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/**<u></u>mailman/listinfo/nek5000-users</a><u></u><<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><u></u>><br>
or, via email, send a message with subject or body 'help' to<br>
nek5000-users-request@lists.**<a href="http://mcs.anl.gov" target="_blank"><u></u>mcs.anl.gov</a><<a href="mailto:nek5000-users-request@lists.mcs.anl.gov" target="_blank">nek5000-users-<u></u>request@lists.mcs.anl.gov</a>><br>
<br>
You can reach the person managing the list at<br>
nek5000-users-owner@lists.mcs.<u></u>**<a href="http://anl.gov" target="_blank">anl.gov</a><<a href="mailto:nek5000-users-owner@lists.mcs.anl.gov" target="_blank">nek5000-users-owner@<u></u>lists.mcs.anl.gov</a>><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of Nek5000-users digest..."<br>
<br>
<br>
Today's Topics:<br>
<br>
1. Re: run-time hang up in gs_setup (nek5000-users@lists.mcs.anl.*<u></u>*gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>><br>
)<br>
<br>
<br>
------------------------------<u></u>**----------------------------<u></u>--**<br>
----------<br>
<br>
Message: 1<br>
Date: Thu, 12 Jan 2012 18:09:48 +0100<br>
<br>
From: nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
Subject: Re: [Nek5000-users] run-time hang up in gs_setup<br>
To: nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
Message-ID:<br>
<<u></u>CAGTrLsaGQkuowxvfpK3AxvCtgCEk*<u></u>*as9YBJkX024qF2Sh7RcmdA@mail.*<u></u>*<br>
<a href="http://gmail.com" target="_blank">gmail.com</a><<a href="mailto:CAGTrLsaGQkuowxvfpK3AxvCtgCEkas9YBJkX024qF2Sh7RcmdA@mail.gmail.com" target="_blank">CAGTrLsaGQkuowxvfpK3<u></u>AxvCtgCEkas9YBJkX024qF2Sh7Rcmd<u></u>A@mail.gmail.com</a>><br>
><br>
<br>
Content-Type: text/plain; charset=ISO-8859-1<br>
<br>
Hi Azad,<br>
<br>
can you try to run the turbChannel example using AMG and the latest<br>
version of the repo. Let me know if this works for you.<br>
<br>
Stefan<br>
<br>
On 1/11/12, nek5000-users@lists.mcs.anl.**<u></u>gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>><br>
<nek5000-users@lists.mcs.anl.*<u></u>*gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>>><br>
wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Dear Stefan and Aleks;<br>
<br>
Thanks for updating the wiki webpage regarding the AMG, although, I<br>
persume there must be another step also there exist, namely: copy the<br>
generated files from the amg_matlb to the running directory? (Or they<br>
should be remained there and one puts the generated .dat files after<br>
running the 3rd step?). By the way non of the versions I tried working<br>
(even 707!) despite the fact that I had a range of matlab versions<br>
tried. Hanging with the old version I had I compiled again and have<br>
got the four files which was rather fast (with the message at the end:<br>
Error contraction factor: 0.47...) I used them and every time during<br>
the run-time it crashed simply:<br>
##############################<u></u>**############################<u></u>##**<br>
###############<br>
AMG: reading through row 144540, pass 121/121<br>
AMG: reading 0.071106 MB of W<br>
AMG: reading 0.115601 MB of AfP<br>
AMG: reading 0.132477 MB of Aff<br>
AMG level 1 F-vars: 440159<br>
AMG level 2 F-vars: 55146<br>
AMG level 3 F-vars: 28480<br>
AMG level 4 F-vars: 17051<br>
AMG level 5 F-vars: 7524<br>
AMG level 6 F-vars: 5711<br>
AMG level 7 F-vars: 5763<br>
AMG level 8 F-vars: 28583<br>
AMG level 9 F-vars: 5380<br>
AMG level 10 F-vars: 5737<br>
Application 731033 exit codes: 139<br>
Application 731033 exit signals: Killed<br>
Application 731033 resources: utime ~417s, stime ~3s<br>
##############################<u></u>**############################<u></u>##**<br>
##############<br>
<br>
Can you help me with that cause I believe this case still doable with<br>
correct AMG scheme.<br>
<br>
Many thanks<br>
Azad<br>
<br>
<br>
<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Hi Azad,<br>
<br>
I believe old AMG files should work up to and including revision 707<br>
in case >you want to check AMG quickly.<br>
<br>
Best.<br>
Aleks<br>
<br>
<br>
----- Original Message -----<br>
From: nek5000-users at <a href="http://lists.mcs.anl.gov" target="_blank">lists.mcs.anl.gov</a><br>
To: nek5000-users at <a href="http://lists.mcs.anl.gov" target="_blank">lists.mcs.anl.gov</a><br>
Sent: Tuesday, January 10, 2012 10:58:47 AM<br>
Subject: Re: [Nek5000-users] run-time hang up in gs_setup<br>
<br>
Hi Azad,<br>
<br>
your choice of lx1=8 is fine (it's our preferred sweet spot). If you<br>
have a large element count (say > 300'000) the factorization in the<br>
XXt setup phase may take hours. I guess that's why it looks like it's<br>
hanging. Again, there is a known bug which looks the same. So can't<br>
tell exactly what's causing your problem.<br>
<br>
I just updated the Wiki: <a href="https://nek5000.mcs.anl.gov/**" target="_blank">https://nek5000.mcs.anl.gov/**</a><br>
index.php/Amg_matlab <<a href="https://nek5000.mcs.anl.gov/index.php/Amg_matlab" target="_blank">https://nek5000.mcs.anl.gov/<u></u>index.php/Amg_matlab</a>><br>
<br>
Can you verify that it still fails.<br>
<br>
-Stefan<br>
<br>
Quoting nek5000-users-request@lists.**<a href="http://mcs.anl.gov" target="_blank"><u></u>mcs.anl.gov</a><<a href="mailto:nek5000-users-request@lists.mcs.anl.gov" target="_blank">nek5000-users-<u></u>request@lists.mcs.anl.gov</a>><br>
:<br>
<br>
Send Nek5000-users mailing list submissions to<br>
nek5000-users@lists.mcs.anl.**<u></u>gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
<a href="https://lists.mcs.anl.gov/**mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/**<u></u>mailman/listinfo/nek5000-users</a><u></u><<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><u></u>><br>
or, via email, send a message with subject or body 'help' to<br>
nek5000-users-request@lists.**<a href="http://mcs.anl.gov" target="_blank"><u></u>mcs.anl.gov</a><<a href="mailto:nek5000-users-request@lists.mcs.anl.gov" target="_blank">nek5000-users-<u></u>request@lists.mcs.anl.gov</a>><br>
<br>
You can reach the person managing the list at<br>
nek5000-users-owner@lists.mcs.<u></u>**<a href="http://anl.gov" target="_blank">anl.gov</a><<a href="mailto:nek5000-users-owner@lists.mcs.anl.gov" target="_blank">nek5000-users-owner@<u></u>lists.mcs.anl.gov</a>><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of Nek5000-users digest..."<br>
<br>
<br>
Today's Topics:<br>
<br>
1. Re: run-time hang up in gs_setup (nek5000-users@lists.mcs.anl.*<u></u>*<br>
gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>>)<br>
2. Re: run-time hang up in gs_setup (nek5000-users@lists.mcs.anl.*<u></u>*<br>
gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>>)<br>
3. Re: run-time hang up in gs_setup (nek5000-users@lists.mcs.anl.*<u></u>*<br>
gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>>)<br>
<br>
<br>
------------------------------<u></u>**----------------------------<u></u>--**<br>
----------<br>
<br>
Message: 1<br>
Date: Tue, 10 Jan 2012 06:01:29 -0600 (CST)<br>
From: nek5000-users@lists.mcs.anl.**<u></u>gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>><br>
Subject: Re: [Nek5000-users] run-time hang up in gs_setup<br>
To: nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
Message-ID: <Pine.LNX.4.64.1201100557250.*<u></u>*<a href="mailto:6026@v8.mcs.anl.gov" target="_blank">6026@v8.mcs.anl.gov</a><<a href="mailto:Pine.LNX.4.64.1201100557250.6026@v8.mcs.anl.gov" target="_blank">Pine.LNX.<u></u>4.64.1201100557250.6026@v8.<u></u>mcs.anl.gov</a>><br>
><br>
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed<br>
<br>
<br>
Hi Azad,<br>
<br>
You are in record-setting territory for element counts! :)<br>
<br>
Are you using the amg-based coarse-grid solver?<br>
It is certain that you will need to do this (and,<br>
therefore, you will need matlab to process the AMG<br>
operators). There is some discussion of the steps<br>
on the wiki page. We can walk you through this process<br>
if you have any questions.<br>
<br>
What value of lx1 are you using?<br>
<br>
I would recommend fewer elements and a higher value of lx1.<br>
I think it will be easier to manage the data, etc.<br>
<br>
Paul<br>
<br>
<br>
<br>
<br>
<br>
On Tue, 10 Jan 2012, nek5000-users@lists.mcs.anl.**<u></u>gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>>wrote:<br>
<br>
Dear NEKs;<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
I am trying to run a simulation of a turbulent flow in a straight pipe<br>
in high Reynolds number (Re_tau = 1000). After generating the grid with<br>
PRENEK and extrude it using n2to3, the mesh ended up with 4,495,920<br>
elements. It compiled properly; however, trying to run it, hanged up in<br>
the last stage:<br>
##############################<u></u>**############################<u></u>##**<br>
############<br>
<br>
verify mesh topology<br>
-1.000000000000000 1.000000000000000 Xrange<br>
-1.000000000000000 1.000000000000000 Yrange<br>
0.000000000000000 25.00000000000000 Zrange<br>
done :: verify mesh topology<br>
<br>
E-solver strategy: 1 itr<br>
mg_nx: 1 3<br>
mg_ny: 1 3<br>
mg_nz: 1 3<br>
call usrsetvert<br>
done :: usrsetvert<br>
<br>
gs_setup: 866937 unique labels shared<br>
pairwise times (avg, min, max): 0.000241442 0.00019722 0.000265908<br>
crystal router : 0.000458177 0.000445795 0.000471807<br>
used all_to_all method: pairwise<br>
setupds time 5.6048E-02 seconds 1 2 4565612 4495920<br>
setvert3d: 4 86046564 122013924 86046564 86046564<br>
call usrsetvert<br>
done :: usrsetvert<br>
<br>
gs_setup: 8041169 unique labels shared<br>
pairwise times (avg, min, max): 0.00050716 0.000427103 0.00056479<br>
crystal router : 0.0040165 0.00392921 0.00411811<br>
used all_to_all method: pairwise<br>
setupds time 1.0465E+00 seconds <a href="tel:2%204%2086046564" value="+12486046564" target="_blank">2 4 86046564</a> 4495920<br>
setup h1 coarse grid, nx_crs= 2<br>
call usrsetvert<br>
done :: usrsetvert<br>
<br>
gs_setup: 866937 unique labels shared<br>
pairwise times (avg, min, max): 0.000233683 0.000197816 0.00024941<br>
crystal router : 0.000466869 0.00045588 0.000478101<br>
used all_to_all method: pairwise<br>
##############################<u></u>**############################<u></u>##**<br>
############<br>
<br>
<br>
I was wondering if you could help me with that. I attached the run<br>
logfile and also genmap.out.<br>
<br>
Many thanks<br>
Azad<br>
<br>
<br>
</blockquote>
<br>
------------------------------<br>
<br>
Message: 2<br>
Date: Tue, 10 Jan 2012 13:35:22 +0100<br>
From: nek5000-users@lists.mcs.anl.**<u></u>gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>><br>
Subject: Re: [Nek5000-users] run-time hang up in gs_setup<br>
To: nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
Message-ID:<br>
<<u></u>CAGTrLsaexkteQN1Y1NQ3FYz7Q2ab*<u></u>*b5YSLzOv+zeTwdvYXpD3Fw@mail.*<u></u>*<br>
<a href="http://gmail.com" target="_blank">gmail.com</a><<a href="mailto:CAGTrLsaexkteQN1Y1NQ3FYz7Q2abb5YSLzOv%252BzeTwdvYXpD3Fw@mail.gmail.com" target="_blank">CAGTrLsaexkteQN1Y1NQ<u></u>3FYz7Q2abb5YSLzOv%<u></u>2BzeTwdvYXpD3Fw@mail.gmail.com</a><u></u>><br>
><br>
Content-Type: text/plain; charset=ISO-8859-1<br>
<br>
Hi Azad,<br>
<br>
We have seen similar situations. I think this has to do with a known<br>
bug. Unfortunately this bug is hard to reproduce and we haven't<br>
managed to fix it yet.<br>
<br>
-Stefan<br>
<br>
On 1/10/12, nek5000-users@lists.mcs.anl.**<u></u>gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>><br>
<nek5000-users@lists.mcs.anl.*<u></u>*gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>>><br>
wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
Hi Azad,<br>
<br>
You are in record-setting territory for element counts! :)<br>
<br>
Are you using the amg-based coarse-grid solver?<br>
It is certain that you will need to do this (and,<br>
therefore, you will need matlab to process the AMG<br>
operators). There is some discussion of the steps<br>
on the wiki page. We can walk you through this process<br>
if you have any questions.<br>
<br>
What value of lx1 are you using?<br>
<br>
I would recommend fewer elements and a higher value of lx1.<br>
I think it will be easier to manage the data, etc.<br>
<br>
Paul<br>
<br>
<br>
<br>
<br>
<br>
On Tue, 10 Jan 2012, nek5000-users@lists.mcs.anl.**<u></u>gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>>wrote:<br>
<br>
Dear NEKs;<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
I am trying to run a simulation of a turbulent flow in a straight pipe<br>
in high Reynolds number (Re_tau = 1000). After generating the grid<br>
with<br>
PRENEK and extrude it using n2to3, the mesh ended up with 4,495,920<br>
elements. It compiled properly; however, trying to run it, hanged up<br>
in<br>
the last stage:<br>
##############################<u></u>**############################<u></u>##**<br>
############<br>
<br>
verify mesh topology<br>
-1.000000000000000 1.000000000000000 Xrange<br>
-1.000000000000000 1.000000000000000 Yrange<br>
0.000000000000000 25.00000000000000 Zrange<br>
done :: verify mesh topology<br>
<br>
E-solver strategy: 1 itr<br>
mg_nx: 1 3<br>
mg_ny: 1 3<br>
mg_nz: 1 3<br>
call usrsetvert<br>
done :: usrsetvert<br>
<br>
gs_setup: 866937 unique labels shared<br>
pairwise times (avg, min, max): 0.000241442 0.00019722 0.000265908<br>
crystal router : 0.000458177 0.000445795 0.000471807<br>
used all_to_all method: pairwise<br>
setupds time 5.6048E-02 seconds 1 2 4565612 4495920<br>
setvert3d: 4 86046564 122013924 86046564 86046564<br>
call usrsetvert<br>
done :: usrsetvert<br>
<br>
gs_setup: 8041169 unique labels shared<br>
pairwise times (avg, min, max): 0.00050716 0.000427103 0.00056479<br>
crystal router : 0.0040165 0.00392921 0.00411811<br>
used all_to_all method: pairwise<br>
setupds time 1.0465E+00 seconds <a href="tel:2%204%2086046564" value="+12486046564" target="_blank">2 4 86046564</a> 4495920<br>
setup h1 coarse grid, nx_crs= 2<br>
call usrsetvert<br>
done :: usrsetvert<br>
<br>
gs_setup: 866937 unique labels shared<br>
pairwise times (avg, min, max): 0.000233683 0.000197816 0.00024941<br>
crystal router : 0.000466869 0.00045588 0.000478101<br>
used all_to_all method: pairwise<br>
##############################<u></u>**############################<u></u>##**<br>
############<br>
<br>
<br>
I was wondering if you could help me with that. I attached the run<br>
logfile and also genmap.out.<br>
<br>
Many thanks<br>
Azad<br>
<br>
______________________________<u></u>**_________________<br>
</blockquote>
Nek5000-users mailing list<br>
Nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:Nek5000-users@lists.mcs.anl.gov" target="_blank">Nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
<a href="https://lists.mcs.anl.gov/**mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/**<u></u>mailman/listinfo/nek5000-users</a><u></u><<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><u></u>><br>
<br>
<br>
</blockquote>
<br>
------------------------------<br>
<br>
Message: 3<br>
Date: Tue, 10 Jan 2012 17:58:47 +0100<br>
From: nek5000-users@lists.mcs.anl.**<u></u>gov<<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.<u></u>anl.gov</a>><br>
Subject: Re: [Nek5000-users] run-time hang up in gs_setup<br>
To: nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:nek5000-users@lists.mcs.anl.gov" target="_blank">nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
Message-ID: <1326214727.2600.282.camel@**<a href="http://damavand.mech.kth.se" target="_blank">d<u></u>amavand.mech.kth.se</a><<a href="mailto:1326214727.2600.282.camel@damavand.mech.kth.se" target="_blank">1326214727<u></u>.2600.282.camel@damavand.mech.<u></u>kth.se</a>><br>
><br>
Content-Type: text/plain; charset="UTF-8"<br>
<br>
Dear Paul and Stefan;<br>
<br>
Thanks very much for looking into it. I use polynomial order 7th<br>
(lx1=8). For the coarse-grid solver I actually used XXt. I also tried to<br>
use AMG, but unfortunately neither v619 nor the latest version could<br>
have compiled its matlab files and always gives me this error (in<br>
matlab/R2011a):<br>
##############################<u></u>**################<br>
...<br>
sparsification tolerance [1e-4]: stol = 0.0001<br>
<br>
------------------------------<u></u>**----------------------------<u></u>--**<br>
------------<br>
Segmentation violation detected at Tue Jan 10 15:56:46 2012<br>
------------------------------<u></u>**----------------------------<u></u>--**<br>
------------<br>
....<br>
Abnormal termination:<br>
Segmentation violation<br>
....<br>
##############################<u></u>**###############<br>
I have been in the web page: "amg_matlab Matlab based tool to generate<br>
AMG solver inputfiles" (<a href="http://nek5000.mcs.anl.gov/**" target="_blank">http://nek5000.mcs.anl.gov/**</a><br>
index.php/Amg_matlab <<a href="http://nek5000.mcs.anl.gov/index.php/Amg_matlab" target="_blank">http://nek5000.mcs.anl.gov/<u></u>index.php/Amg_matlab</a>>)<br>
which gives me an empty link.<br>
<br>
I had an old version of the .dat files needed to run AMG, which I tried<br>
those as (amg_Aff.dat, amgdmp_i.dat, amg.dat, amg_AfP.dat, amgdmp_p.dat,<br>
amgdmp_j.dat, amg_W.dat) and I have got this error:<br>
<br>
##############################<u></u>**##############<br>
...<br>
AMG: reading through row 142800, pass 119/121<br>
AMG: reading through row 144000, pass 120/121<br>
AMG: reading through row 144540, pass 121/121<br>
ERROR (proc<br>
0000,<br>
/afs/<a href="http://pdc.kth.se/home/a/**anoorani/codes/latest_nek/**" target="_blank">pdc.kth.se/home/a/**<u></u>anoorani/codes/latest_nek/**</a><br>
nek5_svn/trunk/nek/jl/amg.c:**<u></u>468<<a href="http://pdc.kth.se/home/a/anoorani/codes/latest_nek/nek5_svn/trunk/nek/jl/amg.c:468" target="_blank">http://pdc.kth.se/home/a/<u></u>anoorani/codes/latest_nek/<u></u>nek5_svn/trunk/nek/jl/amg.c:<u></u>468</a>><br>
):<br>
AMG: missing data for some<br>
rows<br>
<br>
call exitt: dying ...<br>
##############################<u></u>**##############<br>
<br>
I think AMG could be a possibility to overcome this problem, though I<br>
could not manage to get a run with that one. I look into the problem<br>
with higher polynomial order to see if it reduces the number of elements<br>
dramatically, or at least resolve this issue.<br>
<br>
Best regards<br>
Azad<br>
<br>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%<u></u>**%%%%%%%%%%%%%%%%%%%%%%%%%<br>
<br>
<blockquote class="gmail_quote" style="margin:0pt 0pt 0pt 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Hi Azad,<br>
<br>
We have seen similar situations. I think this has to do with a known<br>
bug. Unfortunately this bug is hard to reproduce and we haven't<br>
managed to fix it yet.<br>
<br>
-Stefan<br>
<br>
On 1/10/12, nek5000-users at <a href="http://lists.mcs.anl.gov" target="_blank">lists.mcs.anl.gov</a><br>
<nek5000-users at <a href="http://lists.mcs.anl.gov" target="_blank">lists.mcs.anl.gov</a>> wrote:<br>
<br>
<br>
Hi Azad,<br>
<br>
You are in record-setting territory for element counts! :)<br>
<br>
Are you using the amg-based coarse-grid solver?<br>
It is certain that you will need to do this (and,<br>
therefore, you will need matlab to process the AMG<br>
operators). There is some discussion of the steps<br>
on the wiki page. We can walk you through this process<br>
if you have any questions.<br>
<br>
What value of lx1 are you using?<br>
<br>
I would recommend fewer elements and a higher value of lx1.<br>
I think it will be easier to manage the data, etc.<br>
<br>
Paul<br>
<br>
<br>
<br>
<br>
<br>
On Tue, 10 Jan 2012, nek5000-users at <a href="http://lists.mcs.anl.gov" target="_blank">lists.mcs.anl.gov</a> wrote:<br>
<br>
Dear NEKs;<br>
<br>
I am trying to run a simulation of a turbulent flow in a straight pipe<br>
in high Reynolds number (Re_tau = 1000). After generating the grid with<br>
PRENEK and extrude it using n2to3, the mesh ended up with 4,495,920<br>
elements. It compiled properly; however, trying to run it, hanged up in<br>
the last stage:<br>
##############################<u></u>**############################<u></u>##**<br>
############<br>
<br>
verify mesh topology<br>
-1.000000000000000 1.000000000000000 Xrange<br>
-1.000000000000000 1.000000000000000 Yrange<br>
0.000000000000000 25.00000000000000 Zrange<br>
done :: verify mesh topology<br>
<br>
E-solver strategy: 1 itr<br>
mg_nx: 1 3<br>
mg_ny: 1 3<br>
mg_nz: 1 3<br>
call usrsetvert<br>
done :: usrsetvert<br>
<br>
gs_setup: 866937 unique labels shared<br>
pairwise times (avg, min, max): 0.000241442 0.00019722 0.000265908<br>
crystal router : 0.000458177 0.000445795 0.000471807<br>
used all_to_all method: pairwise<br>
setupds time 5.6048E-02 seconds 1 2 4565612 4495920<br>
setvert3d: 4 86046564 122013924 86046564 86046564<br>
call usrsetvert<br>
done :: usrsetvert<br>
<br>
gs_setup: 8041169 unique labels shared<br>
pairwise times (avg, min, max): 0.00050716 0.000427103 0.00056479<br>
crystal router : 0.0040165 0.00392921 0.00411811<br>
used all_to_all method: pairwise<br>
setupds time 1.0465E+00 seconds <a href="tel:2%204%2086046564" value="+12486046564" target="_blank">2 4 86046564</a> 4495920<br>
setup h1 coarse grid, nx_crs= 2<br>
call usrsetvert<br>
done :: usrsetvert<br>
<br>
gs_setup: 866937 unique labels shared<br>
pairwise times (avg, min, max): 0.000233683 0.000197816 0.00024941<br>
crystal router : 0.000466869 0.00045588 0.000478101<br>
used all_to_all method: pairwise<br>
##############################<u></u>**############################<u></u>##**<br>
############<br>
<br>
<br>
I was wondering if you could help me with that. I attached the run<br>
logfile and also genmap.out.<br>
<br>
Many thanks<br>
Azad<br>
<br>
<br>
</blockquote>
<br>
<br>
------------------------------<br>
<br>
______________________________<u></u>**_________________<br>
Nek5000-users mailing list<br>
Nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:Nek5000-users@lists.mcs.anl.gov" target="_blank">Nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
<a href="https://lists.mcs.anl.gov/**mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/**<u></u>mailman/listinfo/nek5000-users</a><u></u><<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><u></u>><br>
<br>
<br>
End of Nek5000-users Digest, Vol 35, Issue 5<br>
******************************<u></u>****************<br>
<br>
<br>
</blockquote>
<br>
<br>
------------------------------<u></u>**----------------------------<u></u>--**----<br>
This message was sent using IMP, the Internet Messaging Program.<br>
______________________________<u></u>**_________________<br>
Nek5000-users mailing list<br>
Nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:Nek5000-users@lists.mcs.anl.gov" target="_blank">Nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
<a href="https://lists.mcs.anl.gov/**mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/**<u></u>mailman/listinfo/nek5000-users</a><u></u><<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><u></u>><br>
<br>
<br>
</blockquote>
<br>
------------------------------<br>
<br>
______________________________<u></u>**_________________<br>
Nek5000-users mailing list<br>
Nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:Nek5000-users@lists.mcs.anl.gov" target="_blank">Nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
<a href="https://lists.mcs.anl.gov/**mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/**<u></u>mailman/listinfo/nek5000-users</a><u></u><<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><u></u>><br>
<br>
<br>
End of Nek5000-users Digest, Vol 35, Issue 7<br>
******************************<u></u>****************<br>
<br>
<br>
</blockquote>
<br>
<br>
------------------------------<u></u>**----------------------------<u></u>--**----<br>
This message was sent using IMP, the Internet Messaging Program.<br>
______________________________<u></u>**_________________<br>
Nek5000-users mailing list<br>
Nek5000-users@lists.mcs.anl.**<u></u>gov <<a href="mailto:Nek5000-users@lists.mcs.anl.gov" target="_blank">Nek5000-users@lists.mcs.anl.<u></u>gov</a>><br>
<a href="https://lists.mcs.anl.gov/**mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/**<u></u>mailman/listinfo/nek5000-users</a><u></u><<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><u></u>><br>
<br>
</blockquote>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <<a href="http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20120112/b0da596a/attachment.htm" target="_blank">http://lists.mcs.anl.gov/<u></u>pipermail/nek5000-users/<u></u>attachments/20120112/b0da596a/<u></u>attachment.htm</a>><br>
<br>
------------------------------<br>
<br>
______________________________<u></u>_________________<br>
Nek5000-users mailing list<br>
<a href="mailto:Nek5000-users@lists.mcs.anl.gov" target="_blank">Nek5000-users@lists.mcs.anl.<u></u>gov</a><br>
<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><br>
<br>
<br>
End of Nek5000-users Digest, Vol 35, Issue 9<br>
******************************<u></u>**************<br>
<br>
</blockquote>
<br>
<br>
<br>
------------------------------<u></u>------------------------------<u></u>----<br>
This message was sent using IMP, the Internet Messaging Program.<br>
______________________________<u></u>_________________<br>
Nek5000-users mailing list<br>
<a href="mailto:Nek5000-users@lists.mcs.anl.gov" target="_blank">Nek5000-users@lists.mcs.anl.<u></u>gov</a><br>
<a href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users" target="_blank">https://lists.mcs.anl.gov/<u></u>mailman/listinfo/nek5000-users</a><br>
</blockquote></div><br>