[Trilinos-Users] Static condensation

[Erik Boman]

Jesse,

Just to expand on Mike's answer: The Amesos/MUMPS interface is scalable 
but we do not currently support the explicit computation of the Schur 
complement through MUMPS (or any other direct solver). You can compute 
the Schur complement yourself in Epetra (using the factored top left 
block from Amesos/MUMPS) but it will be a bit less efficient.

Regards,
Erik

Heroux, Michael A wrote:
> Jesse,
>
> I am unaware of a turnkey solution for this approach, although the 
> pieces are there to do the "heavy lifting" in a distributed memory 
> environment, if you take the time to learn Epetra's import/export 
> functionality.  I don't know about MUMPS' capabilities in this area.
>
> Mike
>
> From: Jesse Chan <jchan985 at gmail.com <mailto:jchan985 at gmail.com>>
> Date: Sunday, August 12, 2012 8:35 PM
> To: "trilinos-users at software.sandia.gov 
> <mailto:trilinos-users at software.sandia.gov>" 
> <trilinos-users at software.sandia.gov 
> <mailto:trilinos-users at software.sandia.gov>>
> Subject: [EXTERNAL] [Trilinos-Users] Static condensation
>
> Hi all
> Is there an existing way in Trilinos to perform distributed static 
> condensation or compute Schur complements and modified RHS?  Is the 
> Amesos MUMPS routine scalable?  
>
> Thanks
> - Jesse
> ------------------------------------------------------------------------
>
> _______________________________________________
> Trilinos-Users mailing list
> Trilinos-Users at software.sandia.gov
> http://software.sandia.gov/mailman/listinfo/trilinos-users
>