Skip to content

Conversation

@rcknaus
Copy link
Contributor

@rcknaus rcknaus commented Feb 27, 2025

Currently diffs and needs some correctness testing. But the idea is to glue the parallel_op and periodic_op together, as a periodic replicated node has almost always the same behavior as a shared node -- when you want a parallel sum, you also want a periodic sum.

@rcknaus rcknaus marked this pull request as draft February 27, 2025 18:44
auto* nodeVol =
meta_.get_field(stk::topology::NODE_RANK, "tioga_nodal_volume");
stk::mesh::parallel_max(bulk_, {nodeVol});
comm::scatter_max(realm_.bulk_data(), {nodeVol});
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I had to make this change to get it to compile

Suggested change
comm::scatter_max(realm_.bulk_data(), {nodeVol});
sierra::nalu::comm::scatter_max(bulk_, {nodeVol});

@sseraj
Copy link
Contributor

sseraj commented Apr 3, 2025

When I try to run the hackathon test case on Kestrel GPUs with these changes, I get many errors during the decomposition like this:

Face Id: 358685 is not properly covered
 Number of nodes connected to this face is: 4
 attached node Id: 35491
 attached node Id: 35726
 attached node Id: 35725
 attached node Id: 35490
 Number of elements connected to this face is: 1
 attached element Id: 35868

@rcknaus
Copy link
Contributor Author

rcknaus commented Apr 3, 2025

When I try to run the hackathon test case on Kestrel GPUs with these changes, I get many errors during the decomposition like this:

Face Id: 358685 is not properly covered
 Number of nodes connected to this face is: 4
 attached node Id: 35491
 attached node Id: 35726
 attached node Id: 35725
 attached node Id: 35490
 Number of elements connected to this face is: 1
 attached element Id: 35868

I'm guessing this is that I failed to properly add the periodic parts back into the part vec for the check_for_missing_bcs option

@marchdf
Copy link
Contributor

marchdf commented Oct 6, 2025

Hi @rcknaus, how do you feel about this PR? Is there more that you want to do?

@github-actions
Copy link

github-actions bot commented Nov 6, 2025

This PR is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 7 days.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants