[Trilinos-Users] Tpetra & block matrices (FEM)

[Cihan Altinay]

I have only just started to add Trilinos to our code base (a finite 
element solver tool) and the goal is to be able to solve sparse linear 
systems. Currently I create a Tpetra CrsGraph, then a CrsMatrix from 
that to solve block size 1 systems iteratively with Belos.

Eventually we will require various preconditioners (Ifpack2), direct 
solvers (Amesos2) and the ability to solve complex systems (Komplex) 
with various block sizes.

Is that going to work?

Regards,
Cihan


On 10/02/16 10:20, Alicia Klinvex wrote:
> Can you describe the problem you're trying to solve and the packages of
> Trilinos you intend to use?  (Since BlockCrs is a newer class, not all
> packages support it yet.)
>
> On Tue, Feb 9, 2016 at 5:11 PM, Cihan Altinay <c.altinay at uq.edu.au
> <mailto:c.altinay at uq.edu.au>> wrote:
>
>     Great! Do you know what the state of the classes is in the latest
>     release (12.4.2), i.e. is it worth for me to switch to using the
>     block classes or should I rather wait? I don't mind if the
>     interfaces change as long as they are functional.
>
>     Thanks again,
>     Cihan
>
>     On 10/02/16 10:08, Alicia Klinvex wrote:
>
>         Hello,
>
>         Tpetra still has block support in the Experimental namespace:
>         https://trilinos.org/docs/dev/packages/tpetra/doc/html/classTpetra_1_1Experimental_1_1BlockMultiVector.html
>
>         We are actively working on BlockCrsMatrix and BlockMultiVector;
>         since
>         they are in the experimental namespace, they may change at any time.
>
>         Best wishes,
>         Alicia
>
>         On Tue, Feb 9, 2016 at 4:58 PM, Cihan Altinay
>         <c.altinay at uq.edu.au <mailto:c.altinay at uq.edu.au>
>         <mailto:c.altinay at uq.edu.au <mailto:c.altinay at uq.edu.au>>> wrote:
>
>              Hello,
>
>              I was wondering if anyone can shed some light on how they
>         are using
>              Trilinos with FE-like methods that have block sizes > 1?
>
>              I can see from the changelog that BlockCrsGraph,
>         BlockMultiVector
>              etc. have been deprecated and removed from Tpetra in
>         Version 12.0
>              but what's the alternative? Do I need to unroll the blocks
>         myself?
>              Wouldn't that be very inefficient for the solvers?
>
>              Thanks,
>              Cihan
>
>              --
>              Cihan Altinay
>
>              Computer Scientist, Centre for Geoscience Computing
>              eResearch Analyst,  Queensland Cyber Infrastructure Foundation
>              The University of Queensland
>              T: +61 7 334 64118 <tel:%2B61%207%20334%2064118>
>         <tel:%2B61%207%20334%2064118> / F: +61 7 334
>              64134 <tel:%2B61%207%20334%2064134>
>
>              _______________________________________________
>              Trilinos-Users mailing list
>         Trilinos-Users at trilinos.org <mailto:Trilinos-Users at trilinos.org>
>         <mailto:Trilinos-Users at trilinos.org
>         <mailto:Trilinos-Users at trilinos.org>>
>         https://trilinos.org/mailman/listinfo/trilinos-users
>
>
>
>
>     --
>     Cihan Altinay
>
>     Computer Scientist, Centre for Geoscience Computing
>     eResearch Analyst,  Queensland Cyber Infrastructure Foundation
>     The University of Queensland
>     T: +61 7 334 64118 <tel:%2B61%207%20334%2064118> / F: +61 7 334
>     64134 <tel:%2B61%207%20334%2064134>
>
>


-- 
Cihan Altinay

Computer Scientist, Centre for Geoscience Computing
eResearch Analyst,  Queensland Cyber Infrastructure Foundation
The University of Queensland
T: +61 7 334 64118 / F: +61 7 334 64134