> I am developing code in the Step-40 like setting - Cahn- > Hilliard Navier Stokes system. I am witnessing very poor > performance of startup for large problems. > For the system with > 1458 dof for Cahn Hilliard > 47000 dofs for Navier Stoke > > I got the following results (4 computer cores)
Debug or release mode? How does that compare to setup for step-40 or step-32 with a similar problem size? It would also be helpful to know, which part of your setup is slow. You can find out by adding more timing sections. > LA::Vector<double> vec_old_solution(dof_handler_nse.n_dofs()); > > VectorTools::interpolate(dof_handler_nse, ZeroFunction<3>(dim + 1), > vec_old_solution); > > old_solution_nse = vec_old_solution; Are you using a serial vector here, interpolate a zero function, and then copy it over? That will be slow of course. You could just write "old_solution_nse=0" instead. -- Timo Heister http://www.math.clemson.edu/~heister/ -- 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.
