<div dir="ltr">So this is a fully 3D problem, or is it a very flat disc? What is the worst aspect ratio (or whatever) of an element, approximately. That is, is this a bad mesh?<div><br></div><div>You might want to start with a simple problem like a cube. The eigen estimates (<span style="font-size:12.8px">Smooth P0: max eigen=1.011119e+01</span>) are huge and they are a lower bound.</div><div><br></div><div>You might also try <span style="font-size:12.8px">-gamg_est_</span><span style="font-size:12.8px;color:rgb(80,0,80)">ksp_</span><font color="#500050"><span style="font-size:12.8px">max_it 50 and see if these eigen estimates go up much. (and use </span></font><span style="font-size:12.8px">-gamg_est_</span><span style="font-size:12.8px;color:rgb(80,0,80)">ksp_type cg</span><font color="#500050"><span style="font-size:12.8px">). GAMG's eigen estimates are working but I use manufacture a seed vector, which is slightly different than what Cheby does.</span></font></div><div><font color="#500050"><span style="font-size:12.8px"><br></span></font></div><div><font color="#500050"><span style="font-size:12.8px">Also, what version of PETSc are you using? It would be best to use git to clone the repository. This would give you maint or master branch which have a fix for the cheby eigen estimator that your version might not have (use -ksp_view and grep for "noisy" to see if you have an up to date version).</span></font></div><div><font color="#500050"><span style="font-size:12.8px"><br></span></font></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Oct 28, 2016 at 10:12 AM, jeremy theler <span dir="ltr"><<a href="mailto:jeremy@seamplex.com" target="_blank">jeremy@seamplex.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><p dir="ltr">I will try these options in a couple of hours (I have to go out now). I forgot to mention that the geometry has revolution symmetry around the z axis (just the geometry, not the problem because it has a non-symmetric temperature distribution).<br>
I am solving with only one proc, there are approx 50k nodes so 150k dofs.<br>
Thanks again.</p><div class="HOEnZb"><div class="h5">
<br><div class="gmail_quote"><div dir="ltr">On Fri, Oct 28, 2016, 11:07 Mark Adams <<a href="mailto:mfadams@lbl.gov" target="_blank">mfadams@lbl.gov</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr" class="m_-8856447009382966713gmail_msg">Also, try solving the problem with a one level iterative method and Chebyshev, like:<div class="m_-8856447009382966713gmail_msg"><br class="m_-8856447009382966713gmail_msg"><div class="m_-8856447009382966713gmail_msg"><span style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg">-ksp_type chebyshev </span><br class="m_-8856447009382966713gmail_msg"></div><div class="m_-8856447009382966713gmail_msg"><div class="m_-8856447009382966713gmail_msg"><span style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg">-pc_type jacobi</span></div><div class="m_-8856447009382966713gmail_msg"><span style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg"><br class="m_-8856447009382966713gmail_msg"></span></div><div class="m_-8856447009382966713gmail_msg"><span style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg">It will take a long time to solve but I just want to see if it has the same error. </span><br class="m_-8856447009382966713gmail_msg"></div></div></div><div class="m_-8856447009382966713gmail_msg"><span style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg"><br class="m_-8856447009382966713gmail_msg"></span></div></div><div class="gmail_extra m_-8856447009382966713gmail_msg"><br class="m_-8856447009382966713gmail_msg"><div class="gmail_quote m_-8856447009382966713gmail_msg">On Fri, Oct 28, 2016 at 10:04 AM, Mark Adams <span dir="ltr" class="m_-8856447009382966713gmail_msg"><<a href="mailto:mfadams@lbl.gov" class="m_-8856447009382966713gmail_msg" target="_blank">mfadams@lbl.gov</a>></span> wrote:<br class="m_-8856447009382966713gmail_msg"><blockquote class="gmail_quote m_-8856447009382966713gmail_msg" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr" class="m_-8856447009382966713gmail_msg">GAMG's eigen estimator worked but the values are very high. You have very low number of equations per processor, is this a thin body? Are the elements badly stretched?<div class="m_-8856447009382966713gmail_msg"><div style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg"><br class="m_-8856447009382966713gmail_msg"></div><div style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg">Do this again with these parameters:</div><div style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg"><br class="m_-8856447009382966713gmail_msg"></div><div style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg">-mg_levels_ksp_type chebyshev</div><div style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg">-mg_levels_esteig_ksp_type cg</div><div style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg">-mg_levels_esteig_ksp_max_it 10</div><div style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg"><div style="color:rgb(0,0,0);display:inline" class="m_-8856447009382966713gmail_msg"></div>-mg_levels_ksp_chebyshev_<wbr>esteig 0,.1,0,1.05</div>-gamg_est_<span style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg">ksp_type cg</span><br class="m_-8856447009382966713gmail_msg"></div><div class="m_-8856447009382966713gmail_msg"><span style="color:rgb(80,0,80);font-size:12.8px" class="m_-8856447009382966713gmail_msg"><br class="m_-8856447009382966713gmail_msg"></span></div></div><div class="m_-8856447009382966713m_-726016101711307352HOEnZb m_-8856447009382966713gmail_msg"><div class="m_-8856447009382966713m_-726016101711307352h5 m_-8856447009382966713gmail_msg"><div class="gmail_extra m_-8856447009382966713gmail_msg"><br class="m_-8856447009382966713gmail_msg"><div class="gmail_quote m_-8856447009382966713gmail_msg">On Fri, Oct 28, 2016 at 9:48 AM, Jeremy Theler <span dir="ltr" class="m_-8856447009382966713gmail_msg"><<a href="mailto:jeremy@seamplex.com" class="m_-8856447009382966713gmail_msg" target="_blank">jeremy@seamplex.com</a>></span> wrote:<br class="m_-8856447009382966713gmail_msg"><blockquote class="gmail_quote m_-8856447009382966713gmail_msg" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span class="m_-8856447009382966713gmail_msg">On Fri, 2016-10-28 at 09:46 -0400, Mark Adams wrote:<br class="m_-8856447009382966713gmail_msg">
> Please run with -info and grep on GAMG.<br class="m_-8856447009382966713gmail_msg">
><br class="m_-8856447009382966713gmail_msg">
</span>[0] PCSetUp_GAMG(): level 0) N=120726, n data rows=3, n data cols=6,<br class="m_-8856447009382966713gmail_msg">
nnz/row (ave)=41, np=1<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGFilterGraph(): 99.904% nnz after filtering, with<br class="m_-8856447009382966713gmail_msg">
threshold 0., 13.7468 nnz ave. (N=40242)<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGCoarsen_AGG(): Square Graph on level 1 of 1 to square<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGProlongator_AGG(): New grid 1894 nodes<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGOptProlongator_AGG(): Smooth P0: max eigen=5.726852e+00<br class="m_-8856447009382966713gmail_msg">
min=1.330683e-01 PC=jacobi<br class="m_-8856447009382966713gmail_msg">
[0] PCSetUp_GAMG(): 1) N=11364, n data cols=6, nnz/row (ave)=196, 1<br class="m_-8856447009382966713gmail_msg">
active pes<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGFilterGraph(): 99.9839% nnz after filtering, with<br class="m_-8856447009382966713gmail_msg">
threshold 0., 32.7656 nnz ave. (N=1894)<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGProlongator_AGG(): New grid 155 nodes<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGOptProlongator_AGG(): Smooth P0: max eigen=1.011119e+01<br class="m_-8856447009382966713gmail_msg">
min=1.832878e-04 PC=jacobi<br class="m_-8856447009382966713gmail_msg">
[0] PCSetUp_GAMG(): 2) N=930, n data cols=6, nnz/row (ave)=196, 1 active<br class="m_-8856447009382966713gmail_msg">
pes<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGFilterGraph(): 100.% nnz after filtering, with<br class="m_-8856447009382966713gmail_msg">
threshold 0., 32.7806 nnz ave. (N=155)<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGProlongator_AGG(): New grid 9 nodes<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGOptProlongator_AGG(): Smooth P0: max eigen=2.116373e+00<br class="m_-8856447009382966713gmail_msg">
min=6.337173e-03 PC=jacobi<br class="m_-8856447009382966713gmail_msg">
[0] PCSetUp_GAMG(): 3) N=54, n data cols=6, nnz/row (ave)=34, 1 active<br class="m_-8856447009382966713gmail_msg">
pes<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGFilterGraph(): 100.% nnz after filtering, with<br class="m_-8856447009382966713gmail_msg">
threshold 0., 5.66667 nnz ave. (N=9)<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGProlongator_AGG(): New grid 2 nodes<br class="m_-8856447009382966713gmail_msg">
[0] PCGAMGOptProlongator_AGG(): Smooth P0: max eigen=1.984549e+00<br class="m_-8856447009382966713gmail_msg">
min=8.582767e-03 PC=jacobi<br class="m_-8856447009382966713gmail_msg">
[0] PCSetUp_GAMG(): 4) N=12, n data cols=6, nnz/row (ave)=12, 1 active<br class="m_-8856447009382966713gmail_msg">
pes<br class="m_-8856447009382966713gmail_msg">
[0] PCSetUp_GAMG(): 5 levels, grid complexity = 1.48586<br class="m_-8856447009382966713gmail_msg">
error: PETSc error 77-0 'Eigen estimator failed: DIVERGED_NANORINF at<br class="m_-8856447009382966713gmail_msg">
<div class="m_-8856447009382966713m_-726016101711307352m_-7413601766853743386HOEnZb m_-8856447009382966713gmail_msg"><div class="m_-8856447009382966713m_-726016101711307352m_-7413601766853743386h5 m_-8856447009382966713gmail_msg">iteration 0'<br class="m_-8856447009382966713gmail_msg">
in /home/gtheler/libs/petsc-3.7.<wbr>4/src/ksp/ksp/impls/cheby/<wbr>cheby.c<br class="m_-8856447009382966713gmail_msg">
KSPSolve_Chebyshev:440<br class="m_-8856447009382966713gmail_msg">
<br class="m_-8856447009382966713gmail_msg">
<br class="m_-8856447009382966713gmail_msg">
<br class="m_-8856447009382966713gmail_msg">
><br class="m_-8856447009382966713gmail_msg">
><br class="m_-8856447009382966713gmail_msg">
><br class="m_-8856447009382966713gmail_msg">
><br class="m_-8856447009382966713gmail_msg">
<br class="m_-8856447009382966713gmail_msg">
<br class="m_-8856447009382966713gmail_msg">
</div></div></blockquote></div><br class="m_-8856447009382966713gmail_msg"></div>
</div></div></blockquote></div><br class="m_-8856447009382966713gmail_msg"></div>
</blockquote></div>
</div></div></blockquote></div><br></div>