Re: [Libmesh-users] size limit for exodus?

[John P. <jwp...@gm...>]

On Thu, Jun 7, 2012 at 9:07 AM, John Peterson <jwp...@gm...> wrote:
> On Thu, Jun 7, 2012 at 8:47 AM, Paul T. Bauman <ptb...@gm...> wrote:
>> On Thu, Jun 7, 2012 at 9:42 AM, robert <rob...@un...> wrote:
>>
>>> is there a size limit for writing exodus files and reading them with
>>> paraview?
>>>
>>
>> To my knowledge, you are only limited by the memory on your computer. I've
>> used ParaView to visualize solutions with 20+ variables and several million
>> elements on my laptop no problem.
>>
>> TetGenIO TG(mesh);
>>> TG.read ("main_ldio.1.ele");
>>> mesh.write("tryfunwrite.e");
>>>
>>
>> The code is incomplete, but you're writing the mesh (which I'm assuming is
>> a Mesh object) and that will be written in ``libMesh" format. If you want
>> Exodus, you need to use an ExodusII object.
>
> No, actually what he's done here is fine for producing an Exodus file.
>  The mesh.write function tries to be smart like that...
>
> I vaguely remember a "big" or "large" option being available for
> ExodusII files, at least when writing from cubit?

This attachment probably won't make it to the list, but for
completeness, it looks like you are limited to about 90M nodes in the
"original" exodusII format, but the "large model" format is good up to
270M nodes.

With only 800k tets, I doubt you were anywhere near this...

-- 
John