Barry and Jed: Thank you very much for all the hints. Now I have a clue where I have to start ...
Thomas Am 12.06.2012 22:47, schrieb Barry Smith: > On Jun 12, 2012, at 3:42 PM, Jed Brown wrote: > >> On Tue, Jun 12, 2012 at 3:35 PM, Barry Smith<bsmith at mcs.anl.gov> wrote: >> G-S and Shur complement type fieldsplit PCs almost always work right so long >> as you have decent preconditioners for the original block(s). Often you >> don't need much of anything to precondition the Schur complement. >> >> Thomas, note that you'll need Schur or some surrogate preconditioner to deal >> with incompressibility. It'll be worth understanding fieldsplit variants >> while working with Navier-Stokes and Cahn-Hilliard separately. Once you >> understand how to "drive" fieldsplit on those systems separately, you can >> put them together. > Very good point Jed. Until you can solve N.S. efficiently (for example > using PCFIELDSPLIT etc) you really cannot put them together. But that > exercise is not "wasted" effort because you will understand fieldsplit and > Schur preconditioners and then extending to C.H will be much easier. > PCFIELDSPLIT is very much a divide and conquer algorithmic approach, for for > divide and conquer to work you need to know how to conquer the pieces. > > Barry > >> >> >> As I said before you can construct a preconditioner for the N.S. block >> by again using a PCFIELDSPLIT on that (recursive use of PCs). >> >> Note that once you have defined your blocks (in the simple case if you >> have all your degrees of freedom collocated (no staggered grid) just set the >> Vec and Mat block size; otherwise you need to construct an IS that defines >> each block) then experimenting is easy and can be done at the command line >> without changing the code. >>
