Dear Roy,
I do not like somebody talking to me like you.
I just send One email after several months of silence ! and get
answer from
you
which is incredibly poor. Who are you ?
What a curious reply from a Scientist ! What a pity !
I am scientist too, probably at least as experimented as you are
in my own
field.
Just google your name or mine on the web and you will see which of
you or me
has been more scientifically productive in the field and for
what we have
been paid for.
You are probably more productive at being nasty.
I sent this email few days ago, and I received a very nice and
productive
reply from
your nice colleague Benjamin. (FInd it below and let the users
list think of
it).
You should probably learn from him.
Tahar Amari
De : [EMAIL PROTECTED]
Objet : Rép : [Libmesh-users] example probs.
Date : 18 novembre 2007 15:27:47 HNEC
À : [EMAIL PROTECTED]
Cc : [email protected]
Hello,
I am looking for the piece of MPI/ParMetis code that distribute the
Tetrahedral Mesh
over the processors and also the associated unknown to "update".
Please note that I am looking for the low level ones coded in Libmesh
and not
the high level ones.
Could you please let me know where I could find those ?
./contrib/parmetis/Lib is probably what you want.
There are number of partitioning schemes in there -- xyzpart.c uses a
space-filling curve key and sorts it in parallel to get a
partitioning of an
abitrary unstructured mesh.
Parmetis does not distribute the mesh for you, though. That is
onoing work
in libMesh, but MeshCommunication::delete_remote_elements() in
src/mesh/mesh_communication.C will give you an idea of how this
might work.
Hello,
I am looking for the piece of MPI/ParMetis code that distribute the
Tetrahedral Mesh
over the processors and also the associated unknown to "update".
Please note that I am looking for the low level ones coded in Libmesh
and not
the high level ones.
Could you please let me know where I could find those ?
./contrib/parmetis/Lib is probably what you want.
There are number of partitioning schemes in there -- xyzpart.c uses a
space-filling curve key and sorts it in parallel to get a
partitioning of an
abitrary unstructured mesh.
Parmetis does not distribute the mesh for you, though. That is
onoing work
in libMesh, but MeshCommunication::delete_remote_elements() in
src/mesh/mesh_communication.C will give you an idea of how this
might work.
--------------------------------------------
T. Amari
Director of Research at CNRS.
Centre de Physique Theorique
Reply
Ecole Polytechnique
91128 Palaiseau Cedex France
tel : 33 1 69 33 47 53
fax: 33 1 69 33 30 08
email: <mailto:[EMAIL PROTECTED]>
URL : http://www.cpht.polytechnique.fr/cpht/amari
Le 19 nov. 07 à 17:42, Roy Stogner a écrit :
On Sun, 18 Nov 2007, Tahar Amari wrote:
I am looking for
No. You're spamming a copy of a question that's already been
answered, attached to a long copy of an unrelated post, addressed
needlessly to multiple mailing lists as well as several people who
either already subscribe to those mailing lists, aren't interested in
your question, or both.
Stop.
---
Roy
---------------------------------------------------------------------
----
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Libmesh-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/libmesh-devel
