[petsc-dev] What might replace ParMetis

[pascal.tremblay at fr.michelin.com]

Have you considered Zoltan? ( http://www.cs.sandia.gov/Zoltan/ )

It is a higher level package that can call PT-Scotch , ParMETIS or Jostle 
while providing extra-fonctionalities for dynamic mesh partition suitable 
for mesh adaptation.

It also handles the migration of application data across processors.

Although it is part of Trillinos, it is available also as a stand alone 
package.

Regards,

Pascal Tremblay
Mécanique Numérique
Michelin, Centre de technologie de Ladoux -CTE/DTO/SIM/ER
63040 Clermont-Ferrand Cedex 09 - FRANCE
Tél. (ext.): 04 73 10 91 63
Tél. (int.): 69 163
 





Matthew Knepley <knepley at gmail.com> 
Envoyé par : <petsc-dev-bounces at mcs.anl.gov>
07/01/2011 22:32
Veuillez répondre à
For users of the development version of PETSc <petsc-dev at mcs.anl.gov>


A
For users of the development version of PETSc <petsc-dev at mcs.anl.gov>
cc

Objet
Re: [petsc-dev] What might replace ParMetis






On Fri, Jan 7, 2011 at 8:56 PM, John R. Cary <cary at txcorp.com> wrote:
Barry mentioned concern that ParMetis is orphaned.

Any ideas on what one might switch to for parallel
unstructured mesh generation?

ParMetis does mesh partitioning. Scotch might be a possible alternative,
but I have no experience with it, especially in parallel.

   Matt
 
Thx...John
-- 
What most experimenters take for granted before they begin their 
experiments is infinitely more interesting than any results to which their 
experiments lead.
-- Norbert Wiener
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20110110/1d87dbd1/attachment.html>