On Thu, 16 Aug 2012, Roman Vetter wrote:

> I'm going to need two different meshes (with different dimension) in one
> simulation. On each of them, a transient system is solved, and they
> mutually interact during each timestep. My perception is that the
> intended way to achieve this is by using two EquationSystems. Is that
> correct?

That's unfortunately correct.

If the meshes had the same dimension, then you could do subdomain-only
variables to put them both in the same Mesh and same EquationSystems,
but we don't support mixed-dimension meshes yet.

> An additional challenge will be parallelization. Serial execution will
> take months in some cases, so I'm going to need to parallelize as much
> as possible. Ideally, both systems should be parallelized such that of a
> total of N processors, the first system is parallelized to (and solved
> on) M processors, while the other N-M processors are handling the other
> system.

> What is the intended way (if any) to do this? Initialize libmesh
> twice, with different MPI communicators?

Yes.  Beware that you're getting into "insufficiently tested"
territory here, though - I only know of two or three people who use
libMesh this way, it was added as a new feature just for them, there
have been related regressions since then that we've fixed, and there
may have been related regressions since then that have not been fixed.
---
Roy

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
Libmesh-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-users

Reply via email to