Skip to content

Opt-in direct I/O reads, and what segment alignment is actually worth #9838

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
joseph-isaacs wants to merge 4 commits into
base: develop
Choose a base branch
from
Draft

Opt-in direct I/O reads, and what segment alignment is actually worth #9838

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
4fd7717
feat(io): opt-in direct I/O reads and block-aligned segment writes
joseph-isaacs Sep 10, 2026
c0c91f4
bench: measure direct I/O reads and segment padding policies
joseph-isaacs Sep 10, 2026
964e353
feat(file): group small consecutive segments into shared blocks
joseph-isaacs Sep 11, 2026
0a5e328
bench: run CI benchmarks with direct I/O and 4 KiB segment alignment
joseph-isaacs Sep 11, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions .github/workflows/pr-bench-runner.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -73,6 +73,11 @@ jobs:
needs: build
if: ${{ !cancelled() && needs.build.result == 'success' }}
timeout-minutes: 120
env:
# BENCHMARK EXPERIMENT -- revert before merging.
# Each benchmark writes its own Vortex files and reads them back in this job.
VORTEX_DIRECT_IO: "1"
VORTEX_SEGMENT_PADDING: "always"
runs-on: >-
${{ github.repository == 'vortex-data/vortex'
&& format('runs-on={0}/runner=bench-dedicated/family=c8gd.metal-24xl/image=ubuntu24-full-arm64-pre-v2/tag={1}{2}', github.run_id, inputs.benchmark_id, github.event.pull_request.head.repo.fork == false && '/extras=s3-cache' || '')
Expand Down
5 changes: 5 additions & 0 deletions .github/workflows/sql-bench-matrix.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -103,6 +103,11 @@ jobs:
FLAT_LAYOUT_INLINE_ARRAY_NODE: "1"
# Makes python output nicer
COLUMNS: 120
# BENCHMARK EXPERIMENT -- revert before merging.
# `prepare-data` and the query runs share this job, so these cover both writing the
# Vortex files and reading them back.
VORTEX_DIRECT_IO: "1"
VORTEX_SEGMENT_PADDING: "always"
strategy:
fail-fast: false
matrix:
Expand Down
2 changes: 2 additions & 0 deletions Cargo.lock
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

4 changes: 4 additions & 0 deletions vortex-bench/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,10 @@ version = { workspace = true }
name = "data-gen"
test = false

[[bin]]
name = "direct-io"
test = false

[lints]
workspace = true

Expand Down
Loading
Loading