On Tue, 7 Feb 2012, Derek Gaston wrote:
In System::project_solution() it looks like we're doing a full localization of
the solution vector! Here is the offending line
of code:
solution->localize(*current_local_solution);
Shouldn't this either be a call to the version of localize that takes the
send_list? Or more likely just call update()...
At first glance I'd say yes - this looks like a huge bug in the
original version of this function. I'll wait for Ben to weigh in,
though.
Shouldn't the (later) introduction of ghosted PetscVectors have
ameliorated the problem? They've got all the send_list information
implicitly built-in to the vector. It seems like it should be
possible to avoid the excess communication even when users don't
provide a send_list. Or is it that they have the send_list
information on sending processors, only on receiving processors, so
sending processors can't pre-filter what they send?
Some of my huge runs are dying here (trying to do a full localization of 400
Million DoFs is not working out!).
We're within 1 OOM of hitting 32-bit int based bugs, huh? Nice.
---
Roy
------------------------------------------------------------------------------
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
_______________________________________________
Libmesh-devel mailing list
Libmesh-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libmesh-devel
