[petsc-dev] Stokes and saddle point solvers in PETSc?

[Barry Smith]

On Jun 14, 2011, at 6:04 PM, Jed Brown wrote:

> On Wed, Jun 15, 2011 at 00:10, Barry Smith <bsmith at mcs.anl.gov> wrote:
> 
> It turns out that this is hard to test because Stokes problems usually have mixed spaces of some sort. The collocated methods are generally not robust. We currently have rather poor support for staggered grids, so constructing reasonable grid hierarchies is still a lot of user effort. I don't know of a simple way to fix this using DMDA. This is part of a larger issue of constructing custom interpolants, which doesn't seem easy at present.
> 

    It seems we should provide a DMSDA that is built specifically for staggered grids, with the correct number of "slots" in the correct "locations", it would have appropriate grid hierarchies and interpolation/restriction. Not terribly hard but a bit of basic plumbing code. I would rather have this then try to "tack on" the current DA a bunch more stuff.



   Barry