[bddcml-users] bddcml : example ?
sistek at math.cas.cz
Tue Jun 25 17:55:14 CEST 2013
Perfect, thanks for letting me know. It's nice it works for you.
May I know what type of problems you aim to solve? If these are somewhat
non-standard and the current BDDCML set-up would not work well for
those, I would be happy to look into it closer together to see what
might be done in order to make BDDCML work efficiently for the target
Feel free to let me know.
On 06/25/2013 05:13 PM, bob wrote:
> Thank you, it works (Mumps 4.10.0 (unpatched) + ParMETIS 3.2.0 + Metis
> poisson_on_cube 40 3 2
> ===========Possion on cube solver===========
> | Solves problem |
> | -Îu = 1 on Î© = [0,1]^3, |
> | u = 0 on Î, |
> | using FEM and the BDDCML solver. |
> Characteristics of the problem :
> number of processors nproc = 27
> number of dimensions ndim = 3
> mesh dimension meshdim = 3
> number of elements global nelem = 1728000
> number of subdomains nsub = 27
> number of nodes global nnod = 1771561
> number of DOF ndof = 1771561
> number of levels nlevels = 2
> number of subdomains in levels = 27 1
> Characteristics of iterational process:
> tolerance of error tol = 1.000000000000000E-006
> maximum number of iterations maxit = 500
> number of incresing residual ndecrmax = 50
> using recycling of Krylov method ? 1
> Initializing BDDCML ...
> Initializing BDDCML done.
> Loading data ...
> Loading data done.
> Preconditioner set-up ...
> Preconditioner set-up done.
> Calling Krylov method ...
> Krylov method done.
> Output of PCG: ==============
> Number of iterations: 9
> Convergence reason: 0
> Condition number: 2.761
> Finalizing BDDCML ...
> Finalizing BDDCML done.
> Solution properties========
> L_2 norm: 0.0201653452
> L_inf norm: 0.0562187802
> R^n norm: 32.8502860859
More information about the bddcml-users