itaps-parallel Mesh Instances

[Mark Miller]

Hi Karen,

So, you brought up Exodus and, as it turns out, file formats and I/O libraries are
the context I am MOST familiar with in thinking about these issues.

In the world I come from (Silo and SAF), it is very common for HPC applications
to store different meshes in the same database (a Silo file or a SAF database). I
don't think it is common for Exodus II because Exodus II doesn't really support
that too naturally if at all. But Silo (and SAF) are all designed to store multiple
meshes in a single, coherent 'database.' And, it has served us well to be able to
do that. Note also, that Silo (and SAF) also supports storing different parts of
the same meshes in a 'database.'

Often, we'll have cases where fundamentally different physics are being run
on fundamentally different discretizations (e.g. meshes) of the same physical world and
both (or more) of these are stored in the same Silo database.

Mark


"Devine, Karen D." wrote:

>
> I do understand that one would like to store different parts of the same
> mesh (as in multiblock, adaptively refined, etc.) in the same database.  But
> I do not think users would store distinct meshes in the same database, any
> more than they would store two distinct meshes in a single Exodus file.

--
Mark C. Miller, Lawrence Livermore National Laboratory
email: mailto:miller86 at llnl.gov
(M/T/W) (925)-423-5901 (!!LLNL BUSINESS ONLY!!)
(Th/F)  (530)-753-8511 (!!LLNL BUSINESS ONLY!!)