[MOAB-dev] A PDE example with MOAB.

[Tim Tautges]

So, as a general philosophical statement, I see most discretization support as outside the scope of MOAB proper.  That's 
in contrast to libMesh, dealII, and a number of larger frameworks.  Some of this is obviously due to my background more 
in mesh generation than numerical solution of PDEs.  But, I also think the two are separable.

Right now, the only options for discretization with MOAB are to do it yourself, like Jed's done, or to use 
Trilinos/Intrepid; the latter link has been made recently as part of a climate data vis project.  I don't think that 
code has been released officially yet, but I'm pretty sure you could have it if you asked (or let me know and I can ask 
for you, either way).  The code was written by Kara Peterson from SNL, who works with the Intrepid developers.

- tim

On 04/13/2011 03:59 PM, Vijay S. Mahadevan wrote:
> Jed,
>
> I will check out the dohp library and the test programs to understand
> MOAB usage for elliptic problems. It looks like parts of the library
> already provide the functionality that I was planning to do for
> example, like figuring out dof numbering, connectivity, evaluating
> basis on an element etc. These are very basic FEM building blocks and
> your library will help refine my understanding.
>
> Thanks,
> Vijay
>
> On Wed, Apr 13, 2011 at 3:43 PM, Jed Brown<jed at 59a2.org>  wrote:
>> On Wed, Apr 13, 2011 at 22:15, Vijay S. Mahadevan<vijay.m at gmail.com>  wrote:
>>>
>>> I will definitely look at NEK as per your suggestion but if they only
>>> use MOAB to read the mesh and convert it to their own mesh data
>>> structures, it beats the purpose of my question. Thanks for the
>>> pointer though.
>>
>> I have a library that does a bit of a hybrid. A key aspect is to retain
>> enough structure to make high-order and composite finite element methods
>> efficient while still having reasonable performance for low order. I don't
>> think it's really what you're looking for because it just pulls out arrays
>> for most tasks (though it doesn't maintain a full mesh data structure on its
>> own), but I use sets and tags for some purposes. Be warned that it is quite
>> immature and since I don't have any users that I know of, I change things
>> pretty arbitrarily. There are also several places that assume hex elements
>> rather than work through the correct API. Let me know if you have questions.
>> https://github.com/jedbrown/dohp
>> MOAB/iMesh is not user-visible in a typical assembly loop, so Dohp is
>> probably irrelevant to you
>> https://github.com/jedbrown/dohp/blob/master/src/fs/tests/ellip.c#L430
>

-- 
================================================================
"You will keep in perfect peace him whose mind is
   steadfast, because he trusts in you."               Isaiah 26:3

              Tim Tautges            Argonne National Laboratory
          (tautges at mcs.anl.gov)      (telecommuting from UW-Madison)
          phone: (608) 263-8485      1500 Engineering Dr.
            fax: (608) 263-4499      Madison, WI 53706