Migrate to new stats falsification rules#8345
Open
gatesn wants to merge 10 commits into
Open
Conversation
AdamGS
added a commit
that referenced
this pull request
Jun 11, 2026
## Summary Adds docs for `StatsRewriteRule` and its functions. Can be considered a follow-up for #8345, but can be individually merged. Signed-off-by: Adam Gutglick <adam@spiraldb.com>
gatesn
force-pushed
the
ngates/public-stats-rewrite-rules
branch
from
033c73f to
5ed375f
Compare
Port file pruning to session stats rewrites Signed-off-by: "Nicholas Gates" <nick@nickgates.com> Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: "Nicholas Gates" <nick@nickgates.com> Signed-off-by: Nicholas Gates <nick@nickgates.com>
gatesn
force-pushed
the
ngates/public-stats-rewrite-rules
branch
from
5ed375f to
e8dd011
Compare
Merging this PR will not alter performance
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | Simulation | chunked_bool_canonical_into[(1000, 10)] |
20.3 µs | 35.4 µs | -42.51% |
| ❌ | Simulation | chunked_varbinview_into_canonical[(1000, 10)] |
176.9 µs | 212.7 µs | -16.83% |
| ❌ | Simulation | chunked_varbinview_canonical_into[(100, 100)] |
273.3 µs | 307.8 µs | -11.18% |
| ⚡ | Simulation | varbinview_large |
131.2 µs | 112.5 µs | +16.64% |
| ⚡ | Simulation | bitwise_not_vortex_buffer_mut[128] |
215.3 ns | 186.1 ns | +15.67% |
| ⚡ | Simulation | bitwise_not_vortex_buffer_mut[1024] |
275.6 ns | 246.4 ns | +11.84% |
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing ngates/public-stats-rewrite-rules (fe12e71) with develop (67a2b22)
Footnotes
-
10 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩
Signed-off-by: "Nicholas Gates" <nick@nickgates.com> Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: "Nicholas Gates" <nick@nickgates.com> Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: "Nicholas Gates" <nick@nickgates.com> Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Nicholas Gates <nick@nickgates.com>
Signed-off-by: Nicholas Gates <nick@nickgates.com>
gatesn
changed the title
Signed-off-by: Nicholas Gates <nick@nickgates.com>
AdamGS
reviewed
Jun 12, 2026
| with: | ||
| sccache: s3 | ||
| - uses: actions/checkout@df4cb1c069e1874edd31b4311f1884172cec0e10 # v6 | ||
| - uses: actions/setup-java@be666c2fcd27ec809703dec50e508c2fdc7f6654 # v5 |
AdamGS
reviewed
Jun 12, 2026
| use crate::scalar::Scalar; | ||
| use crate::scalar_fn::fns::stat::StatFn; | ||
|
|
||
| /// A target that can bind abstract statistics to concrete expressions. |
Contributor
There was a problem hiding this comment.
What does it mean to bind a statistic to an expression? What makes statistics abstract?
Contributor
Author
There was a problem hiding this comment.
What does it mean to bind a statistic to an expression? What makes statistics abstract?
Yeah stats stuff I am intermingling with an expressions change, so I think the "stat" expression becomes explicitly an "Expression::Placeholder", meaning it must be replaced prior to execution.
It's the same logic as our StatCatalog. Except the current StatCatalog means you have to re-run falsification over the entire expression any time your stats come from a different place, e.g. FileStats vs ZoneMap vs ArrayStats.
Here you take the falsified expression, then "bind" the stats from wherever you get them from.
AdamGS
reviewed
Jun 12, 2026
| @@ -115,14 +119,52 @@ impl FileStatsLayoutReader { | |||
| Ok(result.as_bool().value() == Some(true)) | |||
| refs | ||
| } | ||
|
|
||
| fn collect_referenced_stat_field_names(expr: &Expression, refs: &mut HashSet<FieldName>) { |
Contributor
There was a problem hiding this comment.
we have Node implemented for Expression, you should be able to visit the tree using that instead of hand-rolling the recursion
| } | ||
| } | ||
|
|
||
| fn bool_literal(expr: &Expression) -> Option<Option<bool>> { |
| return Ok(None); | ||
| }; | ||
| let required_stats = filter_required_stats(&lowered, binder.required_stats); | ||
| if required_stats.map().is_empty() && !matches!(bool_literal(&lowered), Some(Some(true))) { |
Contributor
There was a problem hiding this comment.
maybe add a comment explaining this if statement?
| available_stats, | ||
| required_stats: Relation::new(), | ||
| }; | ||
| let Some(lowered) = bind_stats(predicate, &mut binder)? else { |
Contributor
There was a problem hiding this comment.
Maybe I'm just stupid - but I think both "lowering" and "bindings" are worth defining and explaining somewhere.
Contributor
|
I think I mostly have a lot of questions 😅 |
Signed-off-by: "Nicholas Gates" <nick@nickgates.com>
github-actions
Bot
removed
the
action/benchmark
Contributor
Polar Signals Profiling ResultsLatest Run
Powered by Polar Signals Cloud |
Contributor
Benchmarks: PolarSignals ProfilingVortex (geomean): 1.060x ➖ How to read Verdict and Engines
datafusion / vortex-file-compressed (1.060x ➖, 0↑ 4↓)
No file size changes detected. |
Contributor
Benchmarks: TPC-H SF=1 on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.922x ➖, 5↑ 0↓)
datafusion / vortex-compact (0.923x ➖, 3↑ 0↓)
datafusion / parquet (0.925x ➖, 4↑ 0↓)
datafusion / arrow (0.911x ➖, 8↑ 0↓)
duckdb / vortex-file-compressed (0.945x ➖, 2↑ 0↓)
duckdb / vortex-compact (0.945x ➖, 1↑ 0↓)
duckdb / parquet (0.946x ➖, 5↑ 0↓)
duckdb / duckdb (0.945x ➖, 2↑ 0↓)
File Size Changes (10 files changed, -0.1% overall, 4↑ 6↓)
Totals:
|
Contributor
Benchmarks: FineWeb NVMeVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.938x ➖, 2↑ 0↓)
datafusion / vortex-compact (0.963x ➖, 1↑ 0↓)
datafusion / parquet (0.967x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.970x ➖, 0↑ 0↓)
duckdb / vortex-compact (1.022x ➖, 1↑ 2↓)
duckdb / parquet (0.975x ➖, 1↑ 0↓)
File Size Changes (1 files changed, -0.0% overall, 0↑ 1↓)
Totals:
|
Contributor
Benchmarks: TPC-DS SF=1 on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.017x ➖, 3↑ 4↓)
datafusion / vortex-compact (1.015x ➖, 2↑ 1↓)
datafusion / parquet (1.016x ➖, 3↑ 3↓)
duckdb / vortex-file-compressed (1.030x ➖, 0↑ 7↓)
duckdb / vortex-compact (1.025x ➖, 0↑ 3↓)
duckdb / parquet (1.013x ➖, 2↑ 0↓)
duckdb / duckdb (1.005x ➖, 1↑ 2↓)
File Size Changes (6 files changed, +0.0% overall, 3↑ 3↓)
Totals:
|
Contributor
🚨🚨🚨❌❌❌ SQL BENCHMARK FAILED ❌❌❌🚨🚨🚨Benchmark |
Contributor
Benchmarks: Statistical and Population GeneticsVerdict: No clear signal (low confidence) How to read Verdict and Engines
duckdb / vortex-file-compressed (0.971x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.975x ➖, 0↑ 0↓)
duckdb / parquet (0.955x ➖, 0↑ 0↓)
File Size Changes (1 files changed, +0.0% overall, 1↑ 0↓)
Totals:
|
Contributor
Benchmarks: FineWeb S3Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.795x ➖, 2↑ 0↓)
datafusion / vortex-compact (1.199x ➖, 0↑ 3↓)
datafusion / parquet (0.953x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.071x ➖, 0↑ 1↓)
duckdb / vortex-compact (1.051x ➖, 0↑ 2↓)
duckdb / parquet (1.078x ➖, 0↑ 0↓)
|
Contributor
Benchmarks: Random AccessVortex (geomean): 0.946x ➖ How to read Verdict and Engines
unknown / unknown (1.005x ➖, 2↑ 1↓)
|
Contributor
Benchmarks: TPC-H SF=10 on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.059x ➖, 0↑ 0↓)
datafusion / vortex-compact (1.050x ➖, 0↑ 1↓)
datafusion / parquet (1.047x ➖, 0↑ 0↓)
datafusion / arrow (1.067x ➖, 0↑ 1↓)
duckdb / vortex-file-compressed (1.034x ➖, 0↑ 0↓)
duckdb / vortex-compact (1.032x ➖, 0↑ 0↓)
duckdb / parquet (1.023x ➖, 0↑ 1↓)
duckdb / duckdb (1.018x ➖, 0↑ 0↓)
File Size Changes (27 files changed, +0.0% overall, 13↑ 14↓)
Totals:
|
Contributor
Benchmarks: Clickbench on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.932x ➖, 7↑ 0↓)
datafusion / parquet (0.935x ➖, 6↑ 1↓)
duckdb / vortex-file-compressed (0.952x ➖, 6↑ 2↓)
duckdb / parquet (0.973x ➖, 0↑ 0↓)
duckdb / duckdb (0.966x ➖, 2↑ 0↓)
File Size Changes (106 files changed, +0.0% overall, 51↑ 55↓)
Totals:
|
Contributor
Benchmarks: CompressionVortex (geomean): 0.997x ➖ How to read Verdict and Engines
unknown / unknown (0.977x ➖, 9↑ 1↓)
|
Contributor
Benchmarks: Appian on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.001x ➖, 0↑ 0↓)
datafusion / parquet (1.010x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.998x ➖, 0↑ 0↓)
duckdb / parquet (0.997x ➖, 0↑ 0↓)
duckdb / duckdb (0.992x ➖, 0↑ 0↓)
File Size Changes (4 files changed, -0.0% overall, 2↑ 2↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=10 on S3Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.203x ➖, 0↑ 8↓)
datafusion / vortex-compact (1.071x ➖, 0↑ 4↓)
datafusion / parquet (1.062x ➖, 4↑ 8↓)
duckdb / vortex-file-compressed (0.948x ➖, 0↑ 0↓)
duckdb / vortex-compact (1.020x ➖, 0↑ 0↓)
duckdb / parquet (1.045x ➖, 0↑ 0↓)
|
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
2 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.
Switch over to using the new rewrite rule registry from the session, instead of using the scalar fn vtable.