[petsc-users] Using HYPRE AMG for only one of the linear eqns

[Matthew Knepley]

On Sat, Dec 24, 2011 at 8:11 AM, TAY wee-beng <zonexo at gmail.com> wrote:

>  Hi,
>
> No I'm not. Will using ptype2 or ksptype2 work?
>

I think you should setup


http://www.mcs.anl.gov/petsc/petsc-dev/docs/manualpages/PC/PCFIELDSPLIT.html

probably using


http://www.mcs.anl.gov/petsc/petsc-dev/docs/manualpages/PC/PCFieldSplitSetIS.html#PCFieldSplitSetIS

     Matt

> Yours sincerely,
>
> TAY wee-beng
>
>
> On 24/12/2011 3:07 PM, Matthew Knepley wrote:
>
> On Sat, Dec 24, 2011 at 7:42 AM, TAY wee-beng <zonexo at gmail.com> wrote:
>
>> Hi,
>>
>> I am solving 4 linear equations in my cfd code. I have declared them as:
>>
>> KSP   ksp,ksp_semi_x,ksp_semi_y,ksp_semi_z
>>
>> PC    pc,pc_semi_x,pc_semi_y,pc_semi_z
>>
>> PCType      ptype
>>
>> KSPType        ksptype
>>
>> However, I only want to solve the 1st eqn via hypre (thru command
>> prompt), and the others via KSPBCGS.
>>
>> I now add:
>>
>> ksp_type richardson -pc_type hypre -pc_type_hypre boomeramg
>>
>> Are all the eqns now solved using hypre?
>>
>> How should I specify in the command prompt?
>
>
>
>  Are you using DMDA? If so, its easy to do PCFIELDSPLIT from the command
> line.
>
>     Matt
>
>>
>>
>> --
>> Yours sincerely,
>>
>> TAY wee-beng
>>
>>
>
>
>  --
> 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
>
>


-- 
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-users/attachments/20111224/92089921/attachment.htm>