Hi everyone I'm playing around with step-57. A 3D grid which is globally refined for 4 times has 4096 cells and 112724 dofs. Solving such a case is unbearably slow on my computer. I suspect it is due to the direct solver for SparseDirectUMFPACK solver for A_inverse in the BlockSchurPreconditioner. So I am trying to replace it with another solver. *My question is which solver and preconditioner should be used for the unsymmetric A (actually A tilde) when it becomes large?*
I attempted to use FGMRES solver without preconditioner to replace Line 269 as following: SolverControl solver_control (stokes_matrix.block(0,0).m(), 1e-6*utmp.l2_norm(), true); SolverFGMRES<Vector<double> > gmres(solver_control); gmres.solve(stokes_matrix.block(0,0), dst.block(0), utmp, PreconditionIdentity()); But it couldn't converge even in 2D. I also tried SolverBicgstab without preconditioning, it did converge but no improvement compared to direct solver was observed in the cases that I ran. Thank you Jie -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
