[petsc-dev] SAMRAI and PETSc

[Boyce Griffith]

I maintain some code which is based on the original SAMRAI-PETSc 
interface provided by SAMRAI --- basically, it creates/configures Vec 
objects which use SAMRAI routines to evaluate the vector operations on 
AMR data.  My code currently hooks together SAMRAI 2.4.4 and PETSc 
(currently stuck at PETSc 3.0 until a bit later in the summer).  FWIW, 
Rich Hornung tells me that a new SAMRAI release should be out later in 
the summer, which will bring with it some fairly major API changes.

There is a public code repository here:

    https://code.nyu.edu/projects/show/ibamr

The library includes generic discretization stuff which is mainly 
relevant for cell-centered finite-volume-type methods.

-- Boyce

On 6/21/10 12:33 PM, Barry Smith wrote:
>
> Dinesh,
>
> Bobby Philip is developing a version of PFLOTRAN that uses SAMRAI with
> PETSc and has been maintaining SAMRAI to work with petsc-dev. I don't
> know how much support he has for this but it would be great if that
> version of SAMRAI could be made available via a revision control system
> and shared by other groups. I also don't know if other people are
> using/developing SAMRAI at the moment. We could easily host a repository
> but don't have the people to maintain it ourselves.
>
> Barry
>
> On Jun 20, 2010, at 4:22 PM, Dinesh K Kaushik wrote:
>
>> Dear Barry,
>>
>> I noticed that SAMRAI works with older versions of PETSc. Is this
>> correct? Some groups at KAUST are interested in using this package for
>> mesh adaptation.
>>
>> Thanks,
>>
>> Dinesh
>