Do you think any kind of instability could arise from the fact that I'm
trying to simulate a stationary problem when actually it should be
transient? I recall that once I've made a project (flow around a cylinder)
where the clue to know if the grid was sufficiently refined to go for the
non-station
Another thing that I've not mentioned earlier is that to validate the
results I've followed two separated approaches one using standart FEM and
the other using mixed hybrid FEM and both presents the same issues but they
actually "agree" for the coarse solutions
On Tue, Oct 21, 2014 at 12:33
>That indicates that the linear solve is not a
>problem. -ksp_monitor_true_residual and/or -ksp_monitor_singular_value
>can give you more confidence that the linear system is solved
>successfully.
I've run through the troubleshooting list and actually got some insights:
- It seems like I have
With this other options it actually works but I suspect the solver is not
really changing since the residual of all iterations of Newton's method are
the same between all solvers (direct or not).
My issue is that the solution is worse whenever I increase the degrees of
freedom, maybe there is some
Result:
[0]PETSC ERROR: - Error Message
[0]PETSC ERROR: No support for this operation for this object type!
[0]PETSC ERROR: Matrix format mpiaij does not have a built-in PETSc LU!
[0]PETSC ERROR:
-
I'm having some issues that could probably be related with conditioning
number and I would like to change the itarative solver to a direct solver
to see if it gives me any better results. Any good advice on what method to
set (and the option on the command line )?
Thanks
On Sun, Oct 19, 2014 at 2:
Hi everyone!
Sorry to bother so much but since the site is down I can't retrieve some
very basic info:
How do I configure the linear solver used by PETSc? Can I do this by
command line? Witch are the options?
Thanks
--
C
