<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
On 11/10/2011 08:00 PM, <a class="moz-txt-link-abbreviated" href="mailto:petsc-users-request@mcs.anl.gov">petsc-users-request@mcs.anl.gov</a> wrote:
<blockquote
cite="mid:mailman.77.1320948021.20376.petsc-users@mcs.anl.gov"
type="cite">
<pre wrap="">how to speed up convergence</pre>
</blockquote>
Dear Jed,<br>
<br>
1. I'm consistent now with the use of PetscsScalar in my code. <big>I
was not!!!!! </big><br>
<br>
2. mat_fd_type ds works better now. Running with:<br>
<br>
mpiexec -n 8 ./hoac cylinder -llf_flux -n_out 2 -end_time 0.4
-implicit -pc_type asm -sub_pc_type ilu
-sub_pc_factor_reuse_ordering -sub_pc_factor_reuse_fill -gl
-ksp_type fgmres -sub_pc_factor_levels 0 -snes_monitor
-snes_converged_reason -ksp_converged_reason -ts_view -ksp_pc_side
right -sub_pc_factor_nonzeros_along_diagonal -dt 1.0e-2 -ts_type
arkimex -ksp_gmres_restart 50 -snes_max_fail 100
-snes_max_linear_solve_fail 100 -ksp_max_it 100 -mat_fd_type ds<br>
<br>
I get:<br>
<br>
Timestep 0: dt = 0.01, T = 0, Res[rho] = 1.44982, Res[rhou] =
6.94003, Res[rhov] = 0.524307, Res[E] = 3.70306, CFL = 10.1859<br>
0 SNES Function norm 3.847597576099e+03 <br>
Linear solve did not converge due to DIVERGED_ITS iterations 100<br>
<big> 1 SNES Function norm 2.993575158449e+03 </big><br>
Linear solve did not converge due to DIVERGED_ITS iterations 100<br>
2 SNES Function norm 2.992570405848e+03 <br>
Linear solve did not converge due to DIVERGED_ITS iterations 100<br>
3 SNES Function norm 2.991073138769e+03 <br>
<br>
<br>
But, running with :<br>
<br>
mpiexec -n 8 ./hoac cylinder -llf_flux -n_out 2 -end_time 0.4
-implicit -pc_type asm -sub_pc_type ilu
-sub_pc_factor_reuse_ordering -sub_pc_factor_reuse_fill -gl
-ksp_type fgmres -sub_pc_factor_levels 0 -snes_monitor
-snes_converged_reason -ksp_converged_reason -ts_view -ksp_pc_side
right -sub_pc_factor_nonzeros_along_diagonal -dt 1.0e-2 -ts_type
arkimex -ksp_gmres_restart 50 -snes_max_fail 100
-snes_max_linear_solve_fail 100 -ksp_max_it 100 <br>
<br>
I get:<br>
<br>
Timestep 0: dt = 0.01, T = 0, Res[rho] = 1.44982, Res[rhou] =
6.94003, Res[rhov] = 0.524307, Res[E] = 3.70306, CFL = 10.1859<br>
0 SNES Function norm 3.847597576099e+03 <br>
Linear solve did not converge due to DIVERGED_ITS iterations 100<br>
<big> 1 SNES Function norm 3.633205436192e+03 </big><br>
Linear solve did not converge due to DIVERGED_ITS iterations 100<br>
2 SNES Function norm 3.623266783173e+03 <br>
Linear solve did not converge due to DIVERGED_ITS iterations 100<br>
3 SNES Function norm 3.621467573805e+03 <br>
<br>
<br>
However as you may see the convergence is very very slow? Any
suggestions?<br>
<br>
Kostas<br>
<br>
<br>
<br>
<br>
<br>
</body>
</html>