On Jan 17, 2013, at 1:27 PM, Derek Gaston wrote:
>
> But... if DTK does what it claims to do... and does it well, it will be worth
> the pain
>
> Derek
Interesting, this is exactly the problem I'm going after with the new
MeshfreeInterpolation business, so it would be nice to come up with
Ben,
I thought this might be the case. I'll try to make the interface agnostic
of DTK or the MeshfreeInterpolation... or maybe the DTK stuff will use
MeshfreeInterpolation... we'll see.
We have plans to use MeshfreeInterpolation internally for some stuff
but we also have to support DTK for t
Quick warning, though: we *also* build our own Trilinos and PETSc
modules, so we know that their common dependencies like MPI & BLAS
stuff are identical. You'll need to be sure there aren't any mixed
versions or implementations in your dependency tree either.
---
Roy
Coolness - thanks Paul!
Derek
On Thu, Jan 17, 2013 at 12:28 PM, Paul T. Bauman wrote:
> On Thu, Jan 17, 2013 at 1:27 PM, Derek Gaston wrote:
>
>
>> So I'm going to be configuring libMesh with both Trilinos and PETSc
>> support simultaneously... that should lead to a few fun issue ;-)
>>
>
> F
On Thu, Jan 17, 2013 at 1:27 PM, Derek Gaston wrote:
> So I'm going to be configuring libMesh with both Trilinos and PETSc
> support simultaneously... that should lead to a few fun issue ;-)
>
FWIW, we build in this configuration all the time, so it should "just work"
(TM) right now.
--
One of the larger pains here is all the dependencies that DTK has. It is
built on top of Trilinos and MOAB... and it uses a ton of stuff from
Trilinos:
Teuchos
Tpetra
Shards
Intrepid
Zoltan
In particular... the Tpetra part is a pain. We currently only wrap Epetra
with libMesh so a big part
On Thu, Jan 17, 2013 at 11:02 AM, Roy Stogner wrote:
> For unions of linear simplicial meshes you can just turn each
> polyhedron into more simplices; not a great idea if it was going to
> effect basis conditioning but just fine if it's only used for
> integration.
>
> But that's not even the real
+1 for this. I have several things in mind that depend on a "union mesh"
and it would be awesome to have that capability on hand whenever I get
around to needing it (or they let me have students...).
Best,
Paul
--
Master
On Thu, 17 Jan 2013, Derek Gaston wrote:
On Thu, Jan 17, 2013 at 10:39 AM, Roy Stogner wrote:
This is the right way to do things in general, but the details get
*real* tricky - does the union mesh end up being made of
nearly-arbitrary polyhedra when the input meshes aren't aligned?
T
On Thu, 17 Jan 2013, Derek Gaston wrote:
The way that it works is that each code feeds DTK its mesh (nodes,
elements and connectivity). DTK then creates a "union" mesh out of
all of the meshes of all of the coupled codes.
This is the right way to do things in general, but the details get
On Thu, Jan 17, 2013 at 10:39 AM, Roy Stogner wrote:
> This is the right way to do things in general, but the details get
> *real* tricky - does the union mesh end up being made of
> nearly-arbitrary polyhedra when the input meshes aren't aligned?
The meshes they showed us didn't end up as polyh
So,
We are being required to work with this package called Data Transfer Kit
(DTK) ( https://github.com/CNERG/DataTransferKit ) for doing loose coupling
and mesh to mesh transfers.
The way that it works is that each code feeds DTK its mesh (nodes, elements
and connectivity). DTK then creates a "
12 matches
Mail list logo
