Minimize default sql benchmarks suite#8465
Merged
Merged
Conversation
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
AdamGS
added
changelog/ci
action/benchmark-sql
github-actions
Bot
removed
the
action/benchmark-sql
Comment on lines
+289
to
+324
| outputs: | ||
| benchmark_matrix: ${{ steps.matrix.outputs.benchmark_matrix }} | ||
| steps: | ||
| - name: Resolve benchmark matrix | ||
| id: matrix | ||
| shell: bash | ||
| run: | | ||
| cat > matrix.json <<'__BENCHMARK_MATRIX__' | ||
| ${{ inputs.benchmark_matrix }} | ||
| __BENCHMARK_MATRIX__ | ||
|
|
||
| case "${{ inputs.benchmark_profile }}" in | ||
| full) | ||
| jq -c "." matrix.json > resolved-matrix.json | ||
| ;; | ||
| base) | ||
| jq -c ' | ||
| map( | ||
| select(.id != "tpch-s3-10" and .subcommand != "appian") | ||
| | .data_formats |= map(select(. != "vortex-compact" and . != "duckdb")) | ||
| | .pr_targets |= map(select(.format != "vortex-compact" and (.engine != "duckdb" or .format != "duckdb"))) | ||
| | .develop_targets |= map(select(.format != "vortex-compact" and (.engine != "duckdb" or .format != "duckdb"))) | ||
| ) | ||
| ' matrix.json > resolved-matrix.json | ||
| ;; | ||
| *) | ||
| echo "::error::Unknown benchmark_profile '${{ inputs.benchmark_profile }}'. Expected 'full' or 'base'." | ||
| exit 1 | ||
| ;; | ||
| esac | ||
|
|
||
| { | ||
| echo "benchmark_matrix<<__BENCHMARK_MATRIX__" | ||
| cat resolved-matrix.json | ||
| echo "__BENCHMARK_MATRIX__" | ||
| } >> "$GITHUB_OUTPUT" |
Contributor
There was a problem hiding this comment.
is this overly complex? should be have a way to write a list of strings and have those be converted into bench run commands?
Contributor
Author
There was a problem hiding this comment.
we have a JSON format, this is a transformation on top of it, I can also use that.
Signed-off-by: Adam Gutglick <adam@spiraldb.com>
Merging this PR will not alter performance
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | Simulation | chunked_bool_canonical_into[(1000, 10)] |
20.3 µs | 34.9 µs | -41.86% |
| ⚡ | Simulation | take_10k_random |
255.8 µs | 197.9 µs | +29.25% |
| ⚡ | Simulation | take_10k_contiguous |
276.3 µs | 218.4 µs | +26.49% |
| ⚡ | Simulation | patched_take_10k_contiguous_patches |
291 µs | 232.3 µs | +25.27% |
| ⚡ | Simulation | patched_take_10k_random |
303 µs | 244.2 µs | +24.08% |
| ⚡ | WallTime | cuda/bitpacked_u8/unpack/3bw[100M] |
353.5 µs | 302.3 µs | +16.93% |
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing adamg/minimize-benchmarks (b93e467) with develop (69ce1ed)
Footnotes
-
1 benchmark was skipped, so the baseline result was used instead. If it was deleted from the codebase, click here and archive it to remove it from the performance reports. ↩
Contributor
Author
|
This PR is cursed CI just keeps getting queued |
robert3005
approved these changes
Jun 18, 2026
github-actions
Bot
removed
the
action/benchmark-sql
Contributor
Polar Signals Profiling ResultsLatest Run
Powered by Polar Signals Cloud |
Contributor
Benchmarks: PolarSignals Profiling (base)Vortex (geomean): 0.967x ➖ How to read Verdict and Engines
datafusion / vortex-file-compressed (0.967x ➖, 0↑ 0↓)
No file size changes detected. |
Contributor
Benchmarks: FineWeb NVMe (base)Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.974x ➖, 1↑ 0↓)
datafusion / parquet (1.000x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.987x ➖, 1↑ 0↓)
duckdb / parquet (1.004x ➖, 0↑ 0↓)
File Size Changes (3 files changed, -46.3% overall, 0↑ 3↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=1 on NVME (base)Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.002x ➖, 0↑ 0↓)
datafusion / parquet (0.874x ✅, 9↑ 0↓)
datafusion / arrow (0.947x ➖, 3↑ 1↓)
duckdb / vortex-file-compressed (1.009x ➖, 0↑ 0↓)
duckdb / parquet (1.026x ➖, 0↑ 2↓)
File Size Changes (17 files changed, -44.4% overall, 3↑ 14↓)
Totals:
|
Contributor
Benchmarks: TPC-DS SF=1 on NVME (base)Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.085x ➖, 0↑ 29↓)
datafusion / parquet (1.090x ➖, 1↑ 32↓)
duckdb / vortex-file-compressed (1.105x ❌, 1↑ 44↓)
duckdb / parquet (1.096x ➖, 0↑ 43↓)
File Size Changes (31 files changed, -43.5% overall, 3↑ 28↓)
Totals:
|
Contributor
Benchmarks: Statistical and Population Genetics (base)Verdict: No clear signal (low confidence) How to read Verdict and Engines
duckdb / vortex-file-compressed (1.001x ➖, 0↑ 0↓)
duckdb / parquet (1.014x ➖, 0↑ 0↓)
File Size Changes (3 files changed, -32.3% overall, 1↑ 2↓)
Totals:
|
Contributor
Benchmarks: FineWeb S3 (base)Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.034x ➖, 0↑ 0↓)
datafusion / parquet (1.408x ❌, 0↑ 7↓)
duckdb / vortex-file-compressed (1.126x ➖, 0↑ 1↓)
duckdb / parquet (1.104x ➖, 0↑ 0↓)
|
Contributor
Benchmarks: Clickbench on NVME (base)Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.997x ➖, 1↑ 3↓)
datafusion / parquet (0.997x ➖, 1↑ 1↓)
duckdb / vortex-file-compressed (1.010x ➖, 0↑ 4↓)
duckdb / parquet (0.997x ➖, 1↑ 0↓)
File Size Changes (201 files changed, -39.1% overall, 45↑ 156↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=10 on NVME (base)Verdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.012x ➖, 0↑ 0↓)
datafusion / parquet (1.000x ➖, 0↑ 0↓)
datafusion / arrow (1.095x ➖, 0↑ 10↓)
duckdb / vortex-file-compressed (1.017x ➖, 0↑ 0↓)
duckdb / parquet (1.018x ➖, 0↑ 0↓)
File Size Changes (47 files changed, -44.5% overall, 11↑ 36↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=1 on S3 (base)Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.470x ❌, 0↑ 13↓)
datafusion / parquet (1.493x ❌, 0↑ 20↓)
duckdb / vortex-file-compressed (1.055x ➖, 0↑ 1↓)
duckdb / parquet (1.128x ➖, 0↑ 2↓)
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
This PR changes the default benchmark suite that runs when adding
action/benchmark-sqlby removing the following benchmarks from the baseline:duckdb:duckdbandvortex-compactrunsThe full run is unchanged, and will still run post-merge or when using the
action/benchmark-sql-fulllabel.Happy to remove more things, but I think these are the least contentious targets we can start with.