On Thu, Jan 15, 2009 at 10:24 AM, Nguyen, Hung V ERDC-ITL-MS <span dir="ltr"><<a href="mailto:Hung.V.Nguyen@usace.army.mil">Hung.V.Nguyen@usace.army.mil</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hello Matt,<br>
<br>
>however I would first check your matrix using -pc_type lu -ksp_type preonly<br>
to make sure its not singular.<br>
<br>
I got the error message below while running with option above. Do I have to<br>
build a matrix with type of seqaij/seqbaij to run with the -pc_type lu<br>
option?</blockquote><div><br>1) Either run on a single process, or<br><br>2) Install a parallel LU, such as SuperLU --download-superlu<br><br> Matt<br> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<br>
Thanks,<br>
<br>
-Hung<br>
<br>
<br>
<br>
np .. 138524<br>
np .. 143882<br>
[0]PETSC ERROR: --------------------- Error Message<br>
------------------------------------<br>
[0]PETSC ERROR: [15]PETSC ERROR: [12]PETSC ERROR: [8]PETSC ERROR:<br>
--------------------- Error Message ------------------------------------<br>
[4]PETSC ERROR: [6]PETSC ERROR: --------------------- Error Message<br>
------------------------------------<br>
[13]PETSC ERROR: --------------------- Error Message<br>
------------------------------------<br>
--------------------- Error Message ------------------------------------<br>
[15]PETSC ERROR: [13]PETSC ERROR: No support for this operation for this<br>
object type!<br>
[12]PETSC ERROR: [15]PETSC ERROR: No support for this operation for this<br>
object type!<br>
No support for this operation for this object type!<br>
[12]PETSC ERROR: Matrix type mpiaij symbolic LU!<br>
Matrix type mpiaij symbolic LU!<br>
[13]PETSC ERROR: [15]PETSC ERROR: [12]PETSC ERROR: Matrix type mpiaij<br>
symbolic LU!<br>
[13]PETSC ERROR:<br>
------------------------------------------------------------------------<br>
<br>
<br>
<br>
-----Original Message-----<br>
From: <a href="mailto:owner-petsc-users@mcs.anl.gov">owner-petsc-users@mcs.anl.gov</a> [mailto:<a href="mailto:owner-petsc-users@mcs.anl.gov">owner-petsc-users@mcs.anl.gov</a>] On<br>
Behalf Of Matthew Knepley<br>
Sent: Wednesday, January 14, 2009 2:05 PM<br>
To: <a href="mailto:petsc-users@mcs.anl.gov">petsc-users@mcs.anl.gov</a><br>
Subject: Re: Stopping criteria<br>
<br>
On Wed, Jan 14, 2009 at 1:54 PM, Nguyen, Hung V ERDC-ITL-MS<br>
<<a href="mailto:Hung.V.Nguyen@usace.army.mil">Hung.V.Nguyen@usace.army.mil</a>> wrote:<br>
<br>
<br>
<br>
Hello All,<br>
<br>
I tried to solve an ill-conditioned system using cg with Jacobi<br>
preconditioned. The KSP solver was stopping due to diverged reason<br>
within a<br>
few iterations. Is there a way to keep KSP solver running until<br>
max_it?<br>
<br>
<br>
There is no way to continue CG here because it gets a zero divisor, and<br>
interprets this as an indefinite matrix. You can try GMRES, however I would<br>
first check your matrix using -pc_type lu -ksp_type preonly to make sure its<br>
not singular.<br>
<br>
Matt<br>
<br>
<br>
<br>
Thanks,<br>
<br>
-hung<br>
<br>
hvnguyen:jade23% aprun -n 16 ./test_matrix_read -ksp_type cg -pc_type<br>
jacobi<br>
-ksp_rtol 1.0e-15 -ksp_max_it 50000 -ksp_monitor<br>
-ksp_converged_reason<br>
0 KSP Residual norm 1.379074550666e+04<br>
1 KSP Residual norm 7.252034661743e+03<br>
2 KSP Residual norm 7.302184771313e+03<br>
3 KSP Residual norm 1.162244351275e+04<br>
4 KSP Residual norm 7.912531765659e+03<br>
5 KSP Residual norm 4.094706251487e+03<br>
6 KSP Residual norm 5.486131070301e+03<br>
7 KSP Residual norm 6.367904529202e+03<br>
8 KSP Residual norm 6.312767173219e+03<br>
Linear solve did not converge due to DIVERGED_INDEFINITE_MAT<br>
iterations 9<br>
Time in PETSc solver: 0.452695 seconds<br>
The number of iteration = 9<br>
The solution residual error = 6.312767e+03<br>
<font color="#888888"><br>
<br>
<br>
<br>
<br>
<br>
--<br>
What most experimenters take for granted before they begin their experiments<br>
is infinitely more interesting than any results to which their experiments<br>
lead.<br>
-- Norbert Wiener<br>
<br>
</font></blockquote></div><br><br clear="all"><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>