[petsc-dev] [petsc-users] PETSc VecLoad not obeying HDF5 group?

[Matthew Knepley]

On Sat, Nov 15, 2014 at 2:56 AM, Jed Brown <jed at jedbrown.org> wrote:

> Barry Smith <bsmith at mcs.anl.gov> writes:
>
> >   Sorry no one answered this before, I had assumed someone who knew
> >   something about HDF5 would answer
>
> Matt, it would be useful if you could write a summary of PETSc's
> assumptions about HDF5.  I have only dealt with the low-level functions
> and I don't know how you envision metadata being encoded and used.  It's
> clear that PETSc provides a simplified view of what an HDF5 file is.
>

Will do it. RIght now it is incredibly simple, and does just enough to make
a
transient simulation (PyLith) work. It should really be extensible so that
it can
support optimization, etc.

There has been a request from Imperial to look at

  https://ice.txcorp.com/trac/vizschema/wiki/WikiStart

which would expand what we deliver in HDF5 metadata.I have only glanced at
it,
but since we work with them, they might be willing to change the things we
do not
like.

   Matt

-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20141115/78f1ab70/attachment.html>