<div dir="ltr"><br><br><div class="gmail_quote"><div dir="ltr">On Mon, Nov 5, 2018 at 4:11 PM Appel, Thibaut <<a href="mailto:t.appel17@imperial.ac.uk">t.appel17@imperial.ac.uk</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="auto">
<div>"Local" as in serial?<br></div></div></blockquote><div><br></div><div>Block Jacobi with ILU as the solver on each block. Each block corresponds to an MPI process by default. So it is completely parallel it is just not a true ILU. I the limit of one equation per processor it is just (point) Jacobi.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"><div>
<br>
Thibaut</div>
<div><br>
On 5 Nov 2018, at 20:12, Mark Adams <<a href="mailto:mfadams@lbl.gov" target="_blank">mfadams@lbl.gov</a>> wrote:<br>
<br>
</div>
<blockquote type="cite">
<div>
<div dir="ltr"><br>
<br>
<div class="gmail_quote">
<div dir="ltr">On Mon, Nov 5, 2018 at 12:50 PM Thibaut Appel <<a href="mailto:t.appel17@imperial.ac.uk" target="_blank">t.appel17@imperial.ac.uk</a>> wrote:<br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF">
<p>Hi Mark,</p>
<p>Yes it doesn't seem to be usable. Unfortunately we're aiming to do 3D so direct solvers are not a viable solution and PETSc' ILU is not parallel and we can't use HYPRE (complex arithmetic)</p>
</div>
</blockquote>
<div><br>
</div>
<div>I think SuperLU has a parallel ILU but in my opinion parallel ILU is not a big deal. Neither is optimal and the math win (faster convergence) with parallel is offset by the cost of synchronization, in some form, for a true parallel ILU. So I think the
PETSc default gmres/(local)ILU is your best option.</div>
<div> </div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF">
<p>Thibaut<br>
</p>
<div class="m_3866321704781140562m_-4760626044520442305moz-cite-prefix">On 01/11/2018 20:42, Mark Adams wrote:<br>
</div>
<blockquote type="cite">
<div dir="ltr"><br>
<br>
<div class="gmail_quote">
<div dir="ltr">On Wed, Oct 31, 2018 at 8:11 PM Smith, Barry F. <<a href="mailto:bsmith@mcs.anl.gov" target="_blank">bsmith@mcs.anl.gov</a>> wrote:<br>
</div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
<br>
> On Oct 31, 2018, at 5:39 PM, Appel, Thibaut via petsc-users <<a href="mailto:petsc-users@mcs.anl.gov" target="_blank">petsc-users@mcs.anl.gov</a>> wrote:<br>
> <br>
> Well yes naturally for the residual but adding -ksp_true_residual just gives<br>
> <br>
> 0 KSP unpreconditioned resid norm 3.583290589961e+00 true resid norm 3.583290589961e+00 ||r(i)||/||b|| 1.000000000000e+00<br>
> 1 KSP unpreconditioned resid norm 0.000000000000e+00 true resid norm 3.583290589961e+00 ||r(i)||/||b|| 1.000000000000e+00<br>
> Linear solve converged due to CONVERGED_ATOL iterations 1<br>
<br>
Very bad stuff is happening in the preconditioner. The preconditioner must have a null space (which it shouldn't have to be a useful preconditioner).<br>
</blockquote>
<div><br>
</div>
<div>Yea, you are far away from an optimal preconditioner for this system. In low frequency (indefinite) Helmholtz is very very hard. Now, something very bad is going on here but even if you fix it standard AMG is not good for these problems. I would use direct
solvers or grind away it with ILU.</div>
<div> </div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
> <br>
> Mark - if that helps - a Poisson equation is used for the pressure so the Helmholtz is the same as for the velocity in the interior.<br>
> <br>
> Thibaut<br>
> <br>
>> Le 31 oct. 2018 à 21:05, Mark Adams <<a href="mailto:mfadams@lbl.gov" target="_blank">mfadams@lbl.gov</a>> a écrit :<br>
>> <br>
>> These are indefinite (bad) Helmholtz problems. Right?<br>
>> <br>
>> On Wed, Oct 31, 2018 at 2:38 PM Matthew Knepley <<a href="mailto:knepley@gmail.com" target="_blank">knepley@gmail.com</a>> wrote:<br>
>> On Wed, Oct 31, 2018 at 2:13 PM Thibaut Appel <<a href="mailto:t.appel17@imperial.ac.uk" target="_blank">t.appel17@imperial.ac.uk</a>> wrote:<br>
>> Hi Mark, Matthew,<br>
>> <br>
>> Thanks for taking the time.<br>
>> <br>
>> 1) You're not suggesting having -fieldsplit_X_ksp_type fgmres for each field, are you?<br>
>> <br>
>> 2) No, the matrix has pressure in one of the fields. Here it's a 2D problem (but we're also doing 3D), the unknowns are (p,u,v) and those are my 3 fields. We are dealing with subsonic/transsonic flows so it is convection dominated indeed.<br>
>> <br>
>> 3) We are in frequency domain with respect to time, i.e. \partial{phi}/\partial{t} = -i*omega*phi.<br>
>> <br>
>> 4) Hypre is unfortunately not an option since we are in complex arithmetic.<br>
>> <br>
>> <br>
>> <br>
>>> I'm not sure about "-fieldsplit_pc_type gamg" GAMG should work on one block, and hence be a subpc. I'm not up on fieldsplit syntax.<br>
>> According to the online manual page this syntax applies the suffix to all the defined fields?<br>
>> <br>
>> <br>
>> <br>
>>> Mark is correct. I wanted you to change the smoother. He shows how to change it to Richardson (make sure you add the self-scale option), which is probably the best choice.<br>
>>> <br>
>>> Thanks,<br>
>>> <br>
>>> Matt<br>
>> <br>
>> You did tell me to set it to GMRES if I'm not mistaken, that's why I tried "-fieldsplit_mg_levels_ksp_type gmres" (mentioned in the email). Also, it wasn't clear whether these should be applied to each block or the whole system, as the online manual pages
+ .pdf manual barely mention smoothers and how to manipulate MG objects with KSP/PC, this especially with PCFIELDSPLIT where examples are scarce.<br>
>> <br>
>> From what I can gather from your suggestions I tried (lines with X are repeated for X={0,1,2})
<br>
>> <br>
>> This looks good. How can an identically zero vector produce a 0 residual? You should always monitor with<br>
>> <br>
>> -ksp_monitor_true_residual.<br>
>> <br>
>> Thanks,<br>
>> <br>
>> Matt <br>
>> -ksp_view_pre -ksp_monitor -ksp_converged_reason \<br>
>> -ksp_type fgmres -ksp_rtol 1.0e-8 \<br>
>> -pc_type fieldsplit \<br>
>> -pc_fieldsplit_type multiplicative \<br>
>> -pc_fieldsplit_block_size 3 \<br>
>> -pc_fieldsplit_0_fields 0 \<br>
>> -pc_fieldsplit_1_fields 1 \<br>
>> -pc_fieldsplit_2_fields 2 \<br>
>> -fieldsplit_X_pc_type gamg \<br>
>> -fieldsplit_X_ksp_type gmres \<br>
>> -fieldsplit_X_ksp_rtol 1e-10 \<br>
>> -fieldsplit_X_mg_levels_ksp_type richardson \<br>
>> -fieldsplit_X_mg_levels_pc_type sor \<br>
>> -fieldsplit_X_pc_gamg_agg_nsmooths 0 \<br>
>> -fieldsplit_X_mg_levels_ksp_richardson_self_scale \<br>
>> -log_view<br>
>> <br>
>> which yields <br>
>> <br>
>> KSP Object: 1 MPI processes<br>
>> type: fgmres<br>
>> restart=30, using Classical (unmodified) Gram-Schmidt Orthogonalization with no iterative refinement<br>
>> happy breakdown tolerance 1e-30<br>
>> maximum iterations=10000, initial guess is zero<br>
>> tolerances: relative=1e-08, absolute=1e-50, divergence=10000.<br>
>> left preconditioning<br>
>> using DEFAULT norm type for convergence test<br>
>> PC Object: 1 MPI processes<br>
>> type: fieldsplit<br>
>> PC has not been set up so information may be incomplete<br>
>> FieldSplit with MULTIPLICATIVE composition: total splits = 3, blocksize = 3<br>
>> Solver info for each split is in the following KSP objects:<br>
>> Split number 0 Fields 0<br>
>> KSP Object: (fieldsplit_0_) 1 MPI processes<br>
>> type: preonly<br>
>> maximum iterations=10000, initial guess is zero<br>
>> tolerances: relative=1e-05, absolute=1e-50, divergence=10000.<br>
>> left preconditioning<br>
>> using DEFAULT norm type for convergence test<br>
>> PC Object: (fieldsplit_0_) 1 MPI processes<br>
>> type not yet set<br>
>> PC has not been set up so information may be incomplete<br>
>> Split number 1 Fields 1<br>
>> KSP Object: (fieldsplit_1_) 1 MPI processes<br>
>> type: preonly<br>
>> maximum iterations=10000, initial guess is zero<br>
>> tolerances: relative=1e-05, absolute=1e-50, divergence=10000.<br>
>> left preconditioning<br>
>> using DEFAULT norm type for convergence test<br>
>> PC Object: (fieldsplit_1_) 1 MPI processes<br>
>> type not yet set<br>
>> PC has not been set up so information may be incomplete<br>
>> Split number 2 Fields 2<br>
>> KSP Object: (fieldsplit_2_) 1 MPI processes<br>
>> type: preonly<br>
>> maximum iterations=10000, initial guess is zero<br>
>> tolerances: relative=1e-05, absolute=1e-50, divergence=10000.<br>
>> left preconditioning<br>
>> using DEFAULT norm type for convergence test<br>
>> PC Object: (fieldsplit_2_) 1 MPI processes<br>
>> type not yet set<br>
>> PC has not been set up so information may be incomplete<br>
>> linear system matrix = precond matrix:<br>
>> Mat Object: 1 MPI processes<br>
>> type: seqaij<br>
>> rows=52500, cols=52500<br>
>> total: nonzeros=1127079, allocated nonzeros=1128624<br>
>> total number of mallocs used during MatSetValues calls =0<br>
>> not using I-node routines<br>
>> 0 KSP Residual norm 3.583290589961e+00 <br>
>> 1 KSP Residual norm 0.000000000000e+00 <br>
>> Linear solve converged due to CONVERGED_ATOL iterations 1<br>
>> <br>
>> so something must not be set correctly. The solution is identically zero everywhere.<br>
>> <br>
>> Is that option list what you meant? If you could let me know what should be corrected.<br>
>> <br>
>> <br>
>> <br>
>> Thanks for your support,<br>
>> <br>
>> <br>
>> <br>
>> Thibaut<br>
>> <br>
>> <br>
>> <br>
>> On 31/10/2018 16:43, Mark Adams wrote:<br>
>>> <br>
>>> <br>
>>> On Tue, Oct 30, 2018 at 5:23 PM Appel, Thibaut via petsc-users <<a href="mailto:petsc-users@mcs.anl.gov" target="_blank">petsc-users@mcs.anl.gov</a>> wrote:<br>
>>> Dear users,<br>
>>> <br>
>>> Following a suggestion from Matthew Knepley I’ve been trying to apply fieldsplit/gamg for my set of PDEs but I’m still encountering issues despite various tests. pc_gamg simply won’t start.<br>
>>> Note that direct solvers always yield the correct, physical result.<br>
>>> Removing the fieldsplit to focus on the gamg bit and trying to solve the linear system on a modest size problem still gives, with<br>
>>> <br>
>>> '-ksp_monitor -ksp_rtol 1.0e-10 -ksp_gmres_restart 300 -ksp_type gmres -pc_type gamg'<br>
>>> <br>
>>> [3]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------<br>
>>> [3]PETSC ERROR: Petsc has generated inconsistent data<br>
>>> [3]PETSC ERROR: Have un-symmetric graph (apparently). Use '-(null)pc_gamg_sym_graph true' to symetrize the graph or '-(null)pc_gamg_threshold -1' if the matrix is structurally symmetric.<br>
>>> <br>
>>> And since then, after adding '-pc_gamg_sym_graph true' I have been getting<br>
>>> [0]PETSC ERROR: --------------------- Error Message --------------------------------------------------------------<br>
>>> [0]PETSC ERROR: Petsc has generated inconsistent data<br>
>>> [0]PETSC ERROR: Eigen estimator failed: DIVERGED_NANORINF at iteration<br>
>>> <br>
>>> -ksp_chebyshev_esteig_noisy 0/1 does not change anything<br>
>>> <br>
>>> Knowing that Chebyshev eigen estimator needs a positive spectrum I tried ‘-mg_levels_ksp_type gmres’ but iterations would just go on endlessly.<br>
>>> <br>
>>> This is OK, but you need to use '-ksp_type fgmres' (this could be why it is failing ...).
<br>
>>> <br>
>>> It looks like your matrix is 1) just the velocity field and 2) very unsymmetric (eg, convection dominated). I would start with ‘-mg_levels_ksp_type richardson -mg_levels_pc_type sor’.<br>
>>> <br>
>>> I would also start with unsmoothed aggregation: '-pc_gamg_nsmooths 0' <br>
>>> <br>
>>> <br>
>>> It seems that I have indeed eigenvalues of rather high magnitude in the spectrum of my operator without being able to determine the reason.<br>
>>> The eigenvectors look like small artifacts at the wall-inflow or wall-outflow corners with zero anywhere else but I do not know how to interpret this.<br>
>>> Equations are time-harmonic linearized Navier-Stokes to which a forcing is applied, there’s no time-marching.<br>
>>> <br>
>>> You mean you are in frequency domain?<br>
>>> <br>
>>> <br>
>>> Matrix is formed with a MPIAIJ type. The formulation is incompressible, in complex arithmetic and the 2D physical domain is mapped to a logically rectangular,<br>
>>> <br>
>>> This kind of messes up the null space that AMG depends on but AMG theory is gone for NS anyway.<br>
>>> <br>
>>> regular collocated grid with a high-order finite difference method.<br>
>>> I determine the ownership of the rows/degrees of freedom of the matrix with PetscSplitOwnership and I’m not using DMDA.<br>
>>> <br>
>>> Our iterative solvers are probably not going to work well on this but you should test hypre also (-pc_type hypre -pc_hypre_type boomeramg). You need to configure PETSc to download hypre.<br>
>>> <br>
>>> Mark<br>
>>> <br>
>>> <br>
>>> The Fortran application code is memory-leak free and has undergone a strict verification/validation procedure for different variations of the PDEs.<br>
>>> <br>
>>> If there’s any problem with the matrix what could help for the diagnostic? At this point I’m running out of ideas so I would really appreciate additional suggestions and discussions.<br>
>>> <br>
>>> Thanks for your continued support,<br>
>>> <br>
>>> <br>
>>> Thibaut<br>
>> <br>
>> <br>
>> -- <br>
>> What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>
>> -- Norbert Wiener<br>
>> <br>
>> <a href="https://www.cse.buffalo.edu/~knepley/" rel="noreferrer" target="_blank">
https://www.cse.buffalo.edu/~knepley/</a><br>
> <br>
<br>
</blockquote>
</div>
</div>
</blockquote>
</div>
</blockquote>
</div>
</div>
</div>
</blockquote>
</div>
</blockquote></div></div>