Skip to content

list_sum benchmarks for bool, run-end, chunked and sparse elements - #11105

Draft
mhk197 wants to merge 1 commit into
developfrom
mk/list-sum-benches
Draft

mhk197 wants to merge 1 commit into
developfrom
mk/list-sum-benches

Conversation

@mhk197

@mhk197 mhk197 commented Nov 2, 2026 •

Copy link
Copy Markdown
Contributor

Summary

Adds list_sum benchmarks over bool, run-end, chunked and sparse elements. list_sum only has a grouped kernel for primitive elements, so these measure the grouped kernels stacked on this PR (#10970, #11106, #11107) without the scan and pushdown that statpopgen needs.

Changes

  • vortex-array/benches/list_sum.rs: bool_sum, run_end_bool_sum, chunked_i32_sum, chunked_bool_sum, chunked_run_end_bool_sum.
  • encodings/sparse/benches/sparse_list_sum.rs: sparse_bool_sum, chunked_sparse_bool_sum. Sparse lives outside vortex-array, so it needs its own bench.
  • Each runs at average list lengths 8, 256 and 4096 (about statpopgen's 4151 genotypes per list), over nullable elements in 4 chunks, with ~1/8 of lists null.

Testing

  • Built and ran both benches locally. No library code changes.

Signed-off-by: Matt Katz <mhkatz97@gmail.com>
@mhk197 mhk197 added the changelog/chore A trivial change label Nov 2, 2026
@mhk197
mhk197 added this pull request to stack #11108 November 2, 2026 17:24

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

changelog/chore A trivial change

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant