perf(duckdb): push down list length expressions#8544
Open
mhk197 wants to merge 3 commits into
Open
Conversation
Merging this PR will improve performance by 11.47%
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | Simulation | chunked_varbinview_into_canonical[(1000, 10)] |
169 µs | 205.5 µs | -17.76% |
| ⚡ | Simulation | copy_nullable[65536] |
1.4 ms | 1 ms | +32.08% |
| ⚡ | Simulation | copy_non_nullable[65536] |
1,089.1 µs | 908.3 µs | +19.91% |
| ⚡ | Simulation | chunked_varbinview_canonical_into[(100, 100)] |
259.7 µs | 224.5 µs | +15.67% |
| ⚡ | Simulation | bitwise_not_vortex_buffer_mut[128] |
244.4 ns | 215.3 ns | +13.55% |
| ⚡ | Simulation | chunked_varbinview_into_canonical[(100, 100)] |
306.5 µs | 271.1 µs | +13.06% |
| ⚡ | Simulation | bitwise_not_vortex_buffer_mut[1024] |
304.7 ns | 275.6 ns | +10.58% |
| 🆕 | Simulation | list_large |
N/A | 10 ms | N/A |
| 🆕 | Simulation | list_medium |
N/A | 144 µs | N/A |
| 🆕 | Simulation | list_small |
N/A | 58.8 µs | N/A |
| 🆕 | Simulation | listview_large |
N/A | 6 ms | N/A |
| 🆕 | Simulation | listview_medium |
N/A | 98.1 µs | N/A |
| 🆕 | Simulation | listview_small |
N/A | 39 µs | N/A |
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing mk/duckdb-list-length-pushdown (42ddb5d) with develop (9567467)
Footnotes
-
4 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. ↩
mhk197
changed the title
mhk197
changed the title
mhk197
changed the title
mhk197
added
action/benchmark-sql
myrrc
reviewed
Jun 25, 2026
myrrc
left a comment
myrrc
left a comment
Contributor
There was a problem hiding this comment.
Overall, PR LGTM sans some small comments.
Please add sqllogictests in slt/duckdb before merging.
In the sqllogic tests, please also add tests where
array length (for same and different columns) is queried both in SELECT and WHERE.
mhk197
added
action/benchmark-sql
mhk197
force-pushed
the
mk/duckdb-list-length-pushdown
branch
from
5f14a76 to
a9e4773
Compare
Signed-off-by: Matt Katz <mhkatz97@gmail.com>
Signed-off-by: Matt Katz <mhkatz97@gmail.com>
mhk197
force-pushed
the
mk/duckdb-list-length-pushdown
branch
from
a9e4773 to
42ddb5d
Compare
github-actions
Bot
removed
the
action/benchmark-sql
Contributor
Polar Signals Profiling ResultsLatest Run
Powered by Polar Signals Cloud |
Contributor
Benchmarks: PolarSignals ProfilingVortex (geomean): 1.045x ➖ How to read Verdict and Engines
datafusion / vortex-file-compressed (1.045x ➖, 0↑ 0↓)
No file size changes detected. |
Contributor
Benchmarks: FineWeb NVMeVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.007x ➖, 0↑ 0↓)
datafusion / parquet (1.000x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (0.979x ➖, 0↑ 0↓)
duckdb / parquet (0.991x ➖, 0↑ 0↓)
File Size Changes (3 files changed, -46.3% overall, 0↑ 3↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=1 on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.975x ➖, 1↑ 0↓)
datafusion / parquet (0.989x ➖, 0↑ 0↓)
datafusion / arrow (1.009x ➖, 1↑ 1↓)
duckdb / vortex-file-compressed (0.973x ➖, 0↑ 0↓)
duckdb / parquet (0.995x ➖, 2↑ 1↓)
File Size Changes (17 files changed, -44.4% overall, 5↑ 12↓)
Totals:
|
Contributor
Benchmarks: TPC-DS SF=1 on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.009x ➖, 0↑ 2↓)
datafusion / parquet (1.003x ➖, 1↑ 0↓)
duckdb / vortex-file-compressed (1.008x ➖, 0↑ 1↓)
duckdb / parquet (1.006x ➖, 2↑ 3↓)
File Size Changes (30 files changed, -43.4% overall, 4↑ 26↓)
Totals:
|
Contributor
Benchmarks: FineWeb S3Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.762x ➖, 1↑ 0↓)
datafusion / parquet (0.780x ➖, 1↑ 0↓)
duckdb / vortex-file-compressed (0.871x ➖, 0↑ 0↓)
duckdb / parquet (0.923x ➖, 0↑ 0↓)
|
Contributor
Benchmarks: Clickbench Sorted on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.937x ➖, 4↑ 1↓)
datafusion / parquet (0.931x ➖, 1↑ 0↓)
duckdb / vortex-file-compressed (1.038x ➖, 0↑ 2↓)
duckdb / parquet (0.981x ➖, 0↑ 0↓)
File Size Changes (201 files changed, -42.6% overall, 55↑ 146↓)
Totals:
|
Contributor
Benchmarks: Statistical and Population GeneticsVerdict: No clear signal (low confidence) How to read Verdict and Engines
duckdb / vortex-file-compressed (0.930x ➖, 2↑ 0↓)
duckdb / parquet (0.939x ➖, 0↑ 0↓)
File Size Changes (3 files changed, -32.3% overall, 1↑ 2↓)
Totals:
|
Contributor
Benchmarks: Clickbench on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.984x ➖, 3↑ 1↓)
datafusion / parquet (1.009x ➖, 0↑ 2↓)
duckdb / vortex-file-compressed (0.980x ➖, 1↑ 0↓)
duckdb / parquet (0.987x ➖, 0↑ 1↓)
File Size Changes (201 files changed, -39.1% overall, 52↑ 149↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=10 on NVMEVerdict: No clear signal (low confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (1.016x ➖, 0↑ 0↓)
datafusion / parquet (1.013x ➖, 0↑ 0↓)
datafusion / arrow (1.037x ➖, 0↑ 0↓)
duckdb / vortex-file-compressed (1.021x ➖, 0↑ 0↓)
duckdb / parquet (1.008x ➖, 0↑ 0↓)
File Size Changes (47 files changed, -44.5% overall, 6↑ 41↓)
Totals:
|
Contributor
Benchmarks: TPC-H SF=1 on S3Verdict: No clear signal (environment too noisy confidence) How to read Verdict and Engines
datafusion / vortex-file-compressed (0.875x ➖, 4↑ 0↓)
datafusion / parquet (0.741x ➖, 8↑ 0↓)
duckdb / vortex-file-compressed (0.910x ➖, 0↑ 0↓)
duckdb / parquet (0.887x ➖, 0↑ 0↓)
|
myrrc
reviewed
Jun 26, 2026
|
|
||
| # Mixing the len/length/array_length aliases across SELECT and WHERE. | ||
| query I | ||
| SELECT len(a) FROM '$__TEST_DIR__/list-length.vortex' |
Contributor
There was a problem hiding this comment.
Let's add EXPLAIN for all queries in the test to test pushdown happened
myrrc
approved these changes
Jun 26, 2026
robert3005
approved these changes
Jun 26, 2026
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.
Pushes DuckDB's list-length scalar function into the Vortex scan as the
list_lengthexpression, so lengths are computed from list offsets/sizes without materializing element values.Pushdowns supported:
SELECT len(list)/length(list)/array_length(list))WHERE array_length(list) >= k, alsolen/length)Each maps to
cast(list_length(col), i64)— DuckDB'slen/array_lengthreturnBIGINTwhilelist_lengthreturnsu64.len/lengthare overloaded with strings/bits, so the filter path needs the argument type to disambiguate. Added a small FFI accessorduckdb_vx_expr_get_return_typeplusExpressionRef::return_type(), and gatelen/length/array_lengthon the bound child beingLIST/ARRAY.Does not currently support
array_length(expr, dim).Stacked on #8495.