[petsc-users] odd SNES behavior

[Barry Smith]

  hypre run 

  Linear solve converged due to CONVERGED_RTOL iterations 8
  9 SNES Function norm 4.819210383260e+04 
Nonlinear solve converged due to CONVERGED_SNORM_RELATIVE iterations 9

  gang run

 9 SNES Function norm 4.895122815462e+04 

  So they are tracking each other well per Newton step: hypre slightly above but also doing more iterations per Newton so probably reasonable.

   It is interesting that the hypre run hits the "small Newton step size" limit, but the gamg does not (this explains why the hypre "takes" less Newton steps).

   You can run with -snes_linesearch_monitor also (my guess is that it is being triggered more with gamg hence more function evaluations).

   Barry


On Mar 6, 2013, at 11:25 AM, "Mark F. Adams" <mark.adams at columbia.edu> wrote:

> I have a SNES solve if I switch the pc_type between gamg and hypre I get some odd results:
> 
> 1) The gamg solve takes two more Newton iterations and converges for a different reason.  This could be OK, but it seems odd.
> 
> 2) The more mystifying issue is that the gamg solve seems to take just a few more total linear solve iterations but does about 5x as many function evaluations.
> 
> I've attached two log files.  Any ideas what is going on here?  It seems like it should be impossible for the choice of PC to effect the nonlinear solve like this.
> 
> Mark
> <log_petsc_gamg><log_petsc_hypre>