Thank you Bruno for your answers, I failed to mention that I use Fenics with C++, not with Python, sorry about that. I just have few points to ask, if you can bear with me.
1 - Is it truly thread-safe? As in running a single analysis, or running >> several (no data sharing), without problems. >> > deal.II is thread-safe but PETSc is not so if you are using deal.II with > PETSc, your code won't be thread-safe. > So can I use solvers as in Tutorial 3, for instance, and I'll be safe and sound? > 2 - Are there facilities of some kind that can help in generating the >> assembly code? In Fenics I just specified the weak formulations. >> > You need to write the code in C++ and loop over cells, quadrature points, > and basis functions yourself. So this will be more verbose than what you do > in Fenics. > I don't mind at all, I started using FEMs way back, when there was no Fenics. However, is someone planning some additions to deal.II to make it less verbose, as in, for example, providing a DSL for that? > 4 - I see Trilinos is included, is it possible to avoid completely >> PETSc/SLEPc, since they're not AFAIK thread-safe? >> > Yes, you can. I am not an expert of SLEPc but I think that the anasazi > package from Trilinos https://trilinos.org/packages/anasazi has similar > capabilities However, I am not sure what is in the Mac binary. I don't know > if it comes with Trilinos and if it does, it may come with only the > packages required to run deal.II. > And just in case, is there a default eigensolver that is shipped for sure with the DMG? > 5 - Is there a multithreaded linear solver and vector operations, or >> should I write them myself? >> > The linear solvers and the vector operations are multithreaded if you use > deal.II own solvers and vectors. Many operations done inside deal.II are > multithreaded but you are in charge of making sure that the code that you > are writing is multithreaded. There is an infrastructure to help you with > that though. > > That is reasonable. This is very good news to me. To what infrastructure are you referring to? > 7 - Is eliminating binary distributions (DMGs and such) on the horizon? >> > No. > > This is really awesome! Thanks, Franco -- 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 dealii+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
