If you make a consistent RHS, then ICC with a positive definite shift might be fine. You could certainly define a dynamic ordering that puts the "good" blocks first. If the transformation can identify the bad blocks, you could solve the embedded problem with PCREDISTRIBUTE. On Jan 9, 2012 6:24 PM, "Geoffrey Irving" <irving at naml.us> wrote:
> On Mon, Jan 9, 2012 at 3:20 PM, Geoffrey Irving <irving at naml.us> wrote: > > Ah, right. I'm using incomplete Cholesky as a preconditioner, which > > usually won't be stable on the subspace. The subspace is derived from > > freezing the normal velocity of points involved in collisions, so it > > has no useful algebraic properties. > > > > It's not too difficult to symbolically apply P to A (it won't change > > the sparsity), but unfortunately that would make the sparsity pattern > > change each iteration, which would significantly increase the cost of > > ICC. > > Sorry, that sentence was poorly phrased. I meant that it's not too > difficult to symbolically eliminate the relevant subspace from A, but > that this would produce a matrix whose size changes each time around. > I could also replace A with PAP directly without changing the sparsity > pattern, but I imagine incomplete Cholesky might choke on the result, > which would be a 3x3 block matrix with a lot of singular diagonal > blocks. > > Geoffrey > -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20120109/cdb4ddd3/attachment-0001.htm>
