dolfin team mailing list archive

[Anders Logg <logg@xxxxxxxxx>]

Yes, it's added now.

Concerning the parallel efforts, parallel input seems to be in good
condition and Ola and I have made some plans for how to fix the
remaining pieces for the parallel mesh. We're planning to distribute a
global numbering for all (required) mesh entities. So for example if
we have piecewise quadratics, each processor will know the global
numbers of all vertices and edges.

With this information available, each processor will be able to
compute its part of the global parallel dofmap by just calling the
same old tabulate_dofs that we call in sequential. I think this is
close to what Niclas Jansson suggested initially and different from
our original plan which was to build a dofmap in parallel and
communicate the dof numberings ("Algorithm 5"). Now we instead
distribute the entity numbering (essentially using "Algorithm 5" but
for entities) from which the dofmap follows.

-- 
Anders


On Thu, Jan 15, 2009 at 06:20:04PM +0000, Garth N. Wells wrote:
> Did you forget to add the demo.py file?
> 
> Garth
> 
> DOLFIN wrote:
> > One or more new changesets pushed to the primary dolfin repository.
> > A short summary of the last three changesets is included below.
> > 
> > changeset:   5413:1e90a4be1a6a63a68df8806d5d60de3e811d5a78
> > tag:         tip
> > user:        Anders Logg <logg@xxxxxxxxx>
> > date:        Thu Dec 18 22:05:22 2008 +0100
> > files:       demo/mesh/partitioning/python/README demo/mesh/partitioning/python/unitsquare.xml.gz dolfin/main/dolfin_main.h dolfin/swig/dolfin_docstrings.i dolfin/swig/dolfin_headers.i site-packages/dolfin/__init__.py
> > description:
> > Implement partitioning demo in Python. Plotting works fine in parallel
> > (each processor creates one plot) but not from C++.
> > 
> > 
> > changeset:   5412:99a95653288af7a6d5bab7a679e3960c15b7789e
> > user:        Anders Logg <logg@xxxxxxxxx>
> > date:        Thu Dec 18 21:45:44 2008 +0100
> > files:       TODO demo/mesh/partitioning/cpp/main.cpp demo/mesh/partitioning/cpp/plotpartitions demo/mesh/partitioning/cpp/unitsquare.xml.gz demo/mesh/partitioning/python/README demo/mesh/partitioning/python/demo.py
> > description:
> > Fix C++ partitioning demo (copied from passembly benchmark).
> > 
> > 
> > changeset:   5411:9d5ee9090bb30283591f889a2f4d2f10bf5c336f
> > user:        Anders Logg <logg@xxxxxxxxx>
> > date:        Thu Dec 18 21:33:38 2008 +0100
> > files:       TODO demo/mesh/intersection/cpp/main.cpp demo/mesh/intersection/python/demo.py
> > description:
> > Consistency updates in intersection demos. Correct in DOLFIN, but bug in Viper
> > (coordinates not updated for mesh).
> > 
> > ----------------------------------------------------------------------
> > For more details, visit http://www.fenics.org/hg/dolfin
> > _______________________________________________
> > DOLFIN-dev mailing list
> > DOLFIN-dev@xxxxxxxxxx
> > http://www.fenics.org/mailman/listinfo/dolfin-dev
> 
> 
> _______________________________________________
> DOLFIN-dev mailing list
> DOLFIN-dev@xxxxxxxxxx
> http://www.fenics.org/mailman/listinfo/dolfin-dev

Attachment: signature.asc
Description: Digital signature