[petsc-users] sor vs. asm + sor

[Matthew Knepley]

On Wed, Sep 5, 2012 at 12:25 PM, Tobin Isaac <tisaac at ices.utexas.edu> wrote:

>
> What's the difference between "-pc_type sor -pc_sor_local_symmetric"
> and "-pc_type asm -sub_pc_type sor -sub_pc_sor_local_symmetric"?
>

The ASM version sticks a Krylov iteration in these by default.

   Matt


> Specifically, this converges in 30 iterations:
>
> ./ex49 -mx 100 -my 100 -elas_ksp_view -elas_ksp_monitor -elas_ksp_type
> cg -elas_pc_type gamg -elas_pc_gamg_verbose 10 -elas_pc_gamg_threshold
> 0. -elas_mg_coarse_pc_type cholesky -elas_pc_mg_smoothup 1
> -elas_pc_mg_smoothdown 1 -elas_mg_levels_ksp_type richardson
> -elas_mg_levels_pc_type asm -elas_ml_levels_1_sub_pc_type sor
>
> While this iterates forever:
>
> ./ex49 -mx 100 -my 100 -elas_ksp_view -elas_ksp_monitor -elas_ksp_type
> cg -elas_pc_type gamg -elas_pc_gamg_verbose 10 -elas_pc_gamg_threshold
> 0. -elas_mg_coarse_pc_type cholesky -elas_pc_mg_smoothup 1
> -elas_pc_mg_smoothdown 1 -elas_mg_levels_ksp_type richardson
> -elas_mg_levels_pc_type sor
>
> Thanks,
> Toby
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
>
> iEYEARECAAYFAlBHiwQACgkQk/TrNolnueXLKACeOtF4ahbmU+moglWAkBn01fCm
> qxEAnjJvvg48mza8TKqgMRDQf6uEtoq+
> =i0vn
> -----END PGP SIGNATURE-----
>
>


-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20120905/d51e6f20/attachment.html>