<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
<br class="">
<div><br class="">
<blockquote type="cite" class="">
<div class="">On 18 Dec 2018, at 12:16, Hapla Vaclav via petsc-dev <<a href="mailto:petsc-dev@mcs.anl.gov" class="">petsc-dev@mcs.anl.gov</a>> wrote:</div>
<br class="Apple-interchange-newline">
<div class=""><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<blockquote type="cite" style="font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
On 17 Dec 2018, at 18:11, Lawrence Mitchell <<a href="mailto:wence@gmx.li" class="">wence@gmx.li</a>> wrote:<br class="">
<br class="">
<br class="">
<blockquote type="cite" class="">On 17 Dec 2018, at 11:56, Hapla Vaclav <<a href="mailto:vaclav.hapla@erdw.ethz.ch" class="">vaclav.hapla@erdw.ethz.ch</a>> wrote:<br class="">
<br class="">
Matt, great that your reminded this email. I actually completely missed it that time.<br class="">
<br class="">
<blockquote type="cite" class="">On 14 Dec 2018, at 19:54, Matthew Knepley via petsc-dev <<a href="mailto:petsc-dev@mcs.anl.gov" class="">petsc-dev@mcs.anl.gov</a>> wrote:<br class="">
</blockquote>
</blockquote>
<br class="">
[...]<br class="">
<br class="">
<blockquote type="cite" class="">
<blockquote type="cite" class="">I would like:<br class="">
<br class="">
- To be able to dump the DMPlex, and fields, on N processes<br class="">
<br class="">
I think the current HDF5 does what you want.<br class="">
<br class="">
- To be able to load the DMPlex, and fields, on P processes. In the first instance, to get things going, I am happy if P=1.<br class="">
<br class="">
I think this also works with arbitrary P, although the testing can be described as extremely thin.<br class="">
</blockquote>
<br class="">
I think we need to be much more precise here. First off, there are now two HDF5 formats:<br class="">
1) PETSC_VIEWER_HDF5_PETSC - store Plex graph serialization<br class="">
2) PETSC_VIEWER_HDF5_XDMF - store XDMF-compatible representation of vertices and cells<br class="">
3) PETSC_VIEWER_HDF5_VIZ slightly extends 2) with some stuff for visualization, you perhaps understand it better<br class="">
<br class="">
PETSC_VIEWER_DEFAULT/PETSC_VIEWER_NATIVE mean store all three above. I think what Lawrence calls Native should be 1).<br class="">
<br class="">
The format 1) is currently written in parallel but loaded sequentially<br class="">
<a href="https://bitbucket.org/petsc/petsc/src/fbb1886742ac2bbe3b4d1df09bff9724d3fee060/src/dm/impls/plex/plexhdf5.c#lines-834" class="">https://bitbucket.org/petsc/petsc/src/fbb1886742ac2bbe3b4d1df09bff9724d3fee060/src/dm/impls/plex/plexhdf5.c#lines-834</a><br class="">
<br class="">
I don't understand, how it can work correctly for a distributed mesh while the Point SF (connecting partitions) is not stored FWICS. I think there's even no PetscSFView_HDF5(). I will check it more deeply soon.<br class="">
<br class="">
The format 2) is for which I implemented parallel DMLoad().<br class="">
Unfortunately, I can't declare it bulletproof until we declare parallel DMPlexInterpolate() as 100% working. I did quite some work towards it in<br class="">
https://bitbucket.org/petsc/petsc/pull-requests/1227/dmplexintepolate-fix-orientation-of-faces/<br class="">
but as stated in the PR summary, there are still some examples failing because of the wrong Point SF which is partly fixed in knepley/fix-plex-interpolate-sf but it seems it's not yet finished. Matt, is there any chance you could look at it at some point in
near future?<br class="">
<br class="">
I think for Lawrence's purposes, 2) can be used to read the initial mesh file but for checkpointing 1) seems to be better ATM because it dumps everything including interpolated edges & faces, labels and perhaps some more additional information.<br class="">
</blockquote>
<br class="">
OK, so I guess there are two different things going on here:<br class="">
<br class="">
1. Store the data you need to reconstruct a DMPlex<br class="">
<br class="">
2. Store the data you need to have a DMPlex viewable via XDMF.<br class="">
<br class="">
3. Store the data you need to reconstruct a DMPlex AND have it viewable via XDMF.<br class="">
<br class="">
For checkpointing only purposes, I only really need 1; for viz purposes, one only needs 2; ideally, one would not separate viz and checkpointing files if there is sufficient overlap of data (I think there is), which needs 3.<br class="">
</blockquote>
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">Yes.
Coordinates can be shared, topology storage is completely different, so with 3, it's just stored in two different ways redundantly.</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">Note
I try to explain it's not that sharp that the XDMF format itself is only for viewing. The format itself is virtually capable to store the whole DMPlex representation. As I say, if you create HDF5 viewer and push PETSC_VIEWER_HDF5_XDMF format, DMLoad() loads
_just_ the XDMF-related data in parallel (using DMPlexCreateFromCellListParallel()). For example, src/dm/impls/plex/examples/tutorials/ex5.c shows that.</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">The
current interface is limited in that it only stores/loads vertices/cells, and edges/faces must be made on-the-fly. However, XDMF itself allows to store also these, so the need to call DMPlexInterpolate() could be eliminated.</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
</div>
</blockquote>
<div><br class="">
</div>
<div>Sorry, this is not entirely correct. Edges/faces are not explicitly stored but can be referred to using cell-local numbering. So labels or fields can still refer to them. But we can't avoid DMPlexInterpolate() after each load. For me it's not such a problem
because parallel DMPlexInterpolate() scales almost perfectly (it's only not entirely correct ATM as I keep reminding).</div>
<div><br class="">
</div>
<div>Vaclav</div>
<br class="">
<blockquote type="cite" class="">
<div class=""><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">There
are also means to store labels using XDMF attributes/sets, and I'm planning to implement this in DMPlex.</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">So
my overall idea (which I presented also at this year's User Meeting in London and nobody has objected yet), is that some FE codes could potentially use only this for both checkpointing and viewing. Advantages would include removing the redundancy in storage
(reducing the file size) and increased interoperability with 3rd party tools.</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">Anyway,
as XDMF refers to HDF5 heavy data, one is free to store any XDMF-unrelated data there as well, which is what happens with PETSC_VIEWER_DEFAULT. The "native PETSc" data, the "PETSc-native" topology, is then stored as well. _At the moment_ it's a good idea because
of the limitations of the _current_ XDMF-HDF5 I/O implementation in PETSc.</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">Of
course, XDMF + corresponding HDF5 part can never be as general as DMPlex itself and the "native PETSc" storage. For instance just because it supports "only" 3 dimensions of a normal world :-) whereas DMPlex support virtually any spatial dimension (a big deal
of functionality relies on spatial dimension <= 3, though).</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">Anyone
knows what e.g. Fenics uses for checkpointing and/or final viewable datafile?</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">Thanks</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<span style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;" class="">Vaclav</span><br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br style="caret-color: rgb(0, 0, 0); font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<blockquote type="cite" style="font-family: Menlo-Regular; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; text-decoration: none;" class="">
<br class="">
<blockquote type="cite" class="">I will nevertheless keep on working to improve 2) so that it can store edges & faces & labels in the XDMF-compatible way.<br class="">
<br class="">
<blockquote type="cite" class=""><br class="">
For dumping, I think I can do DMView(dm) in PETSc "native" format, and that will write out the topology in a global numbering.<br class="">
<br class="">
I would use HDF5.<br class="">
<br class="">
For the field coefficients, I can just VecView(vec). But there does not appear to be any way of saving the Section so that I can actually attach those coefficients to points in the mesh.<br class="">
<br class="">
Hmm, I will check this right now. If it does not exist, I will write it.<br class="">
</blockquote>
<br class="">
No, it certainly doesn't exist. There is only ASCII view implemented.<br class="">
<br class="">
<blockquote type="cite" class=""><br class="">
I can do PetscSectionCreateGlobalSection(section), so that I have a the global numbering for offsets, but presumably for the point numbering, I need to convert the local chart into global point numbers using DMPlexCreatePointNumbering?<br class="">
<br class="">
No, all Sections use local points. We do not use global point numbers anywhere in Plex.<br class="">
</blockquote>
<br class="">
True. DMPlex is partition-wise sequential. The only thing which connects the submeshes is the Point SF.<br class="">
</blockquote>
<br class="">
OK, so I think I misunderstood what the dump format looks like then. For parallel store/load cycle when I go from N to P processes what must I do?<br class="">
<br class="">
If I understand correctly the dump on N processes contains:<br class="">
<br class="">
For each process, in process-local numbering<br class="">
<br class="">
- The DMPlex topology on that process<br class="">
<br class="">
Now, given that the only thing that connects these local pieces of the DM together is the point SF, as Vaclav says, it must be the case that a reloadable dump file contains that information.<br class="">
<br class="">
<br class="">
OK, so to dump a field so that we can reload it we must need:<br class="">
<br class="">
- topology (in local numbering)<br class="">
- point SF (connecting the local pieces of the topology together)<br class="">
- Vector (dofs), presumably in local layout to make things easier<br class="">
- Section describing the vector layout (local numbering)<br class="">
<br class="">
So to load, I do:<br class="">
<br class="">
1. load and distribute topology, and construct new point SF (this presumably gives me a "migration SF" that maps from old points to new points<br class="">
<br class="">
2. Broadcast the Section over migration SF so that we know how many dofs belong to each point in the new topology<br class="">
<br class="">
3. Broadcast the Vec over the migration SF to get the dofs to the right place.<br class="">
<br class="">
Whenever I think of this on paper it seems "easy", but then I occasionally try and sit down and do it and immediately get lost, so I am normally missing something.<br class="">
<br class="">
What am I missing this time?<br class="">
<br class="">
Lawrence</blockquote>
</div>
</blockquote>
</div>
<br class="">
</body>
</html>