Skip to content

1.6.0rc - #231

Merged
LudwigBoess merged 160 commits into
masterfrom
1.6.0rc
Oct 6, 2026
Merged

LudwigBoess merged 160 commits into
masterfrom
1.6.0rc

Conversation

@LudwigBoess

Copy link
Copy Markdown
Collaborator

No description provided.

haykh and others added 30 commits May 18, 2026 08:51
Frontier-specific sorting + general current filter performance
haykh and others added 29 commits September 21, 2026 16:03
Copied byte-identical from dev/qed (277eb74) so a future merge of
that branch resolves these as add/add with matching content.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The restart reader opened the BP file on MPI_COMM_SELF, so every rank
independently parsed the full global metadata instead of a single
rank-0 read + broadcast. Phase 1 also had every rank loop over all
g_ndomains reading each subdomain's extent/ncells via per-element
Mode::Sync Gets. Both costs scale ~O(N_ranks^2) and made restart reads
dominate wall time at large rank counts (~3 h at 65536 ranks vs a
2-3 min write).

- Open the reader on MPI_COMM_WORLD so ADIOS2 reads/broadcasts metadata
  collectively and can aggregate reads.
- In Phase 1, each rank now reads only its own subdomain entry and
  MPI_Allgathers the ncells/xmin/xmax arrays; global_extent and the
  reconstruction check are computed from the gathered layout. Non-MPI
  path unchanged.

Verified ~50x faster on Frontier (read 209 s vs ~3 h) in the sibling
entity_bh tree.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fix slow parallel checkpoint read (COMM_SELF + O(N^2) metadata loop)
@LudwigBoess
LudwigBoess merged commit 62ab34f into master Oct 6, 2026
12 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants