On Mon, 8 Aug 2016, Vikram Garg wrote: > We have a domain split up into two subdomains, the first subdomain > has a higher resolution mesh and solves a complex physics, while the second > has a lower resolution mesh and a simpler physics.
Same variables but with different weak residuals? > We are observing that, with GMRES, Linear problem in both physics? > the completely complex problem solves much faster Fewer iterations? Wall clock time is a lot harder to pin down. What preconditioner are you currently using? > than with a multiphysics setup (10% domain being complex physics). How fast does it solve with 0% domain being complex physics? > We are trying to see if this is due to a linear solver choice > issue. > > Would GMRES have a problem with a subdomain physics/mesh split up ? I'm wondering if rather your physics is more "complex" in terms of residual evaluation but is easier on the solver somehow. > Is PCFIELDSPLIT a better way to go here ? libMesh makes fieldsplits between different variables easy enough now, but I don't think we've added any options to make field splits between subdomains easy, so I think you'll need to code the PETSc index sets yourselves to do this. --- Roy ------------------------------------------------------------------------------ What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic patterns at an interface-level. Reveals which users, apps, and protocols are consuming the most bandwidth. Provides multi-vendor support for NetFlow, J-Flow, sFlow and other flows. Make informed decisions using capacity planning reports. http://sdm.link/zohodev2dev _______________________________________________ Libmesh-users mailing list Libmesh-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/libmesh-users
