Validity casting doesn't hide usage of execution context#7710
Merged
robert3005 merged 9 commits intodevelopfrom Apr 30, 2026
Merged
Validity casting doesn't hide usage of execution context#7710robert3005 merged 9 commits intodevelopfrom
robert3005 merged 9 commits intodevelopfrom
Conversation
Contributor
|
We need to ensure that reduce rule exit early if they cannot match and do little more unless we know they match, since we try to apply them so often. |
robert3005
commented
Apr 29, 2026
Merging this PR will degrade performance by 24.87%
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ⚡ | WallTime | runend[10M_i32_runlen_1000] |
133.1 µs | 96.3 µs | +38.3% |
| ⚡ | WallTime | dynamic_dispatch_f32[10M] |
165.9 µs | 113.7 µs | +45.95% |
| ⚡ | Simulation | take_indices[(1000, 256)] |
38.4 µs | 34.6 µs | +10.91% |
| ⚡ | Simulation | null_count_run_end[(10000, 1024, 0.5)] |
35.4 µs | 32.1 µs | +10.39% |
| ⚡ | Simulation | null_count_run_end[(10000, 1024, 0.1)] |
35.4 µs | 32.1 µs | +10.47% |
| ⚡ | Simulation | null_count_run_end[(10000, 256, 0.5)] |
37 µs | 33.6 µs | +10.09% |
| ❌ | Simulation | new_bp_prim_test_between[i64, 32768] |
177.4 µs | 236.1 µs | -24.87% |
| ⚡ | Simulation | bitwise_not_vortex_buffer_mut[128] |
275.3 ns | 246.1 ns | +11.85% |
Comparing rk/validitycast (55d1554) with develop (e2062a7)
robert3005
force-pushed
the
rk/validitycast
branch
from
77d698b to
9298f56
Compare
Contributor
Author
|
The thing that's extremely repetitive here is |
Signed-off-by: Robert Kruszewski <github@robertk.io>
joseph-isaacs
approved these changes
Apr 30, 2026
Contributor
joseph-isaacs
left a comment
joseph-isaacs
left a comment
There was a problem hiding this comment.
Lets run benchmarks first
robert3005
force-pushed
the
rk/validitycast
branch
from
44dd940 to
74f36ac
Compare
robert3005
added
the
action/benchmark-sql
github-actions
Bot
removed
the
action/benchmark-sql
Contributor
Benchmarks: Clickbench on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.998x ➖, 0↑ 0↓)
datafusion / parquet (1.001x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.997x ➖, 2↑ 1↓)
duckdb / parquet (0.992x ➖, 0↑ 0↓)
duckdb / duckdb (0.982x ➖, 2↑ 1↓)
Full attributed analysis
|
Contributor
Benchmarks: TPC-DS SF=1 on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.987x ➖, 0↑ 2↓)
datafusion / vortex-compact (0.970x ➖, 4↑ 0↓)
datafusion / parquet (0.980x ➖, 2↑ 0↓)
duckdb / vortex-file-compressed (0.981x ➖, 4↑ 0↓)
duckdb / vortex-compact (0.969x ➖, 2↑ 0↓)
duckdb / parquet (0.991x ➖, 0↑ 0↓)
duckdb / duckdb (0.976x ➖, 1↑ 0↓)
Full attributed analysis
|
Contributor
Benchmarks: Statistical and Population GeneticsVerdict: No clear signal (low confidence) duckdb / vortex-file-compressed (1.039x ➖, 0↑ 1↓)
duckdb / vortex-compact (1.009x ➖, 0↑ 0↓)
duckdb / parquet (1.014x ➖, 0↑ 0↓)
Full attributed analysis
|
Contributor
Benchmarks: TPC-H SF=1 on S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.845x ➖, 4↑ 0↓)
datafusion / vortex-compact (0.964x ➖, 0↑ 0↓)
datafusion / parquet (0.963x ➖, 0↑ 1↓)
duckdb / vortex-file-compressed (0.915x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.974x ➖, 0↑ 0↓)
duckdb / parquet (0.968x ➖, 0↑ 0↓)
Full attributed analysis
|
Contributor
Benchmarks: FineWeb NVMeVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (1.032x ➖, 0↑ 0↓)
datafusion / vortex-compact (1.066x ➖, 0↑ 3↓)
datafusion / parquet (1.033x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.035x ➖, 0↑ 0↓)
duckdb / vortex-compact (1.011x ➖, 0↑ 0↓)
duckdb / parquet (1.029x ➖, 0↑ 1↓)
Full attributed analysis
|
Contributor
Benchmarks: TPC-H SF=1 on NVMEVerdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.943x ➖, 0↑ 0↓)
datafusion / vortex-compact (0.947x ➖, 1↑ 0↓)
datafusion / parquet (0.970x ➖, 0↑ 2↓)
datafusion / arrow (0.943x ➖, 2↑ 0↓)
duckdb / vortex-file-compressed (0.953x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.950x ➖, 0↑ 0↓)
duckdb / parquet (0.988x ➖, 0↑ 1↓)
duckdb / duckdb (0.968x ➖, 1↑ 0↓)
Full attributed analysis
|
Contributor
Benchmarks: PolarSignals ProfilingVortex (geomean): 0.983x ➖ datafusion / vortex-file-compressed (0.983x ➖, 1↑ 0↓)
|
Contributor
Benchmarks: FineWeb S3Verdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.982x ➖, 0↑ 0↓)
datafusion / vortex-compact (0.980x ➖, 0↑ 0↓)
datafusion / parquet (0.966x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.992x ➖, 0↑ 0↓)
duckdb / vortex-compact (1.026x ➖, 0↑ 1↓)
duckdb / parquet (0.992x ➖, 0↑ 0↓)
Full attributed analysis
|
Contributor
Benchmarks: TPC-H SF=10 on NVMEVerdict: No clear signal (low confidence) datafusion / vortex-file-compressed (0.985x ➖, 2↑ 0↓)
datafusion / vortex-compact (1.001x ➖, 0↑ 0↓)
datafusion / parquet (0.991x ➖, 0↑ 0↓)
datafusion / arrow (0.982x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.002x ➖, 0↑ 0↓)
duckdb / vortex-compact (1.005x ➖, 0↑ 0↓)
duckdb / parquet (0.996x ➖, 0↑ 0↓)
duckdb / duckdb (0.997x ➖, 0↑ 0↓)
Full attributed analysis
|
Contributor
Benchmarks: TPC-H SF=10 on S3Verdict: No clear signal (environment too noisy confidence) datafusion / vortex-file-compressed (0.868x ➖, 1↑ 1↓)
datafusion / vortex-compact (0.971x ➖, 1↑ 0↓)
datafusion / parquet (0.970x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.964x ➖, 0↑ 0↓)
duckdb / vortex-compact (0.941x ➖, 0↑ 0↓)
duckdb / parquet (0.960x ➖, 0↑ 0↓)
Full attributed analysis
|
robert3005
added
the
action/benchmark-sql
github-actions
Bot
removed
the
action/benchmark-sql
Contributor
Polar Signals Profiling ResultsLatest Run
Powered by Polar Signals Cloud |
Contributor
File Sizes: PolarSignals ProfilingNo file size changes detected. |
Contributor
File Sizes: FineWeb NVMeNo file size changes detected. |
Contributor
File Sizes: TPC-H SF=1 on NVMENo file size changes detected. |
Contributor
File Sizes: TPC-DS SF=1 on NVMENo file size changes detected. |
Contributor
File Sizes: Statistical and Population GeneticsNo file size changes detected. |
Contributor
File Sizes: TPC-H SF=10 on NVMENo file size changes detected. |
Contributor
File Sizes: Clickbench on NVMEFile Size Changes (1 files changed, -0.0% overall, 0↑ 1↓)
Totals:
|
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.
Casting validity correctly advertises that it might do compute by requiring
context.
We split the logic of casting between CastReduce rules which can handle all
NonNullalbe -> Nullable and Nullable -> NonNullable casts if stats have been
calculated and CastKernels which need to handle full generality of casts, in
practice Nullable to NonNullable casts
Signed-off-by: Robert Kruszewski github@robertk.io