Relocate Delta's selection policy into cost-model priors; plumb the model through the builders by connortsui20 · Pull Request #8700 · vortex-data/vortex

1 benchmark regressed

⚡ 11 improved benchmarks
❌ 1 regressed benchmark
✅ 1607 untouched benchmarks
🆕 11 new benchmarks
⏩ 42 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

	Mode	Benchmark	`BASE`	`HEAD`	Efficiency
❌	Simulation	`chunked_varbinview_into_canonical[(1000, 10)]`	169.6 µs	205.8 µs	-17.57%
⚡	Simulation	`rebuild_naive`	109.1 µs	91.1 µs	+19.76%
⚡	Simulation	`pushdown_compare[(1000, 16, 4)]`	150.1 µs	128 µs	+17.2%
⚡	Simulation	`pushdown_compare[(1000, 64, 4)]`	149.9 µs	128.2 µs	+16.92%
⚡	Simulation	`pushdown_compare[(1000, 4, 4)]`	149.6 µs	128.2 µs	+16.72%
⚡	Simulation	`pushdown_compare[(1000, 64, 8)]`	156.2 µs	134.3 µs	+16.26%
⚡	Simulation	`pushdown_compare[(1000, 4, 8)]`	154 µs	133 µs	+15.76%
⚡	Simulation	`pushdown_compare[(1000, 16, 8)]`	162 µs	140.5 µs	+15.34%
⚡	Simulation	`chunked_varbinview_into_canonical[(100, 100)]`	307 µs	271.6 µs	+13.02%
⚡	Simulation	`encode_varbin[(1000, 32)]`	165 µs	147.4 µs	+11.97%
⚡	Simulation	`chunked_varbinview_opt_canonical_into[(100, 100)]`	340.5 µs	305.3 µs	+11.54%
⚡	Simulation	`compare_int_constant`	298 µs	268.7 µs	+10.89%
🆕	Simulation	`fsl_sum_large`	N/A	65.2 ms	N/A
🆕	Simulation	`fsl_sum_medium`	N/A	727.9 µs	N/A
🆕	Simulation	`fsl_sum_small`	N/A	56.8 µs	N/A
🆕	Simulation	`list_sum_large`	N/A	135.7 ms	N/A
🆕	Simulation	`list_sum_medium`	N/A	1.1 ms	N/A
🆕	Simulation	`list_sum_nullable_elements_large`	N/A	495.9 ms	N/A
🆕	Simulation	`list_sum_nullable_elements_medium`	N/A	4.9 ms	N/A
🆕	Simulation	`list_sum_small`	N/A	144.2 µs	N/A
...	...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

_{Comparing cost-model-r5-builders (b81955f) with cost-model-r3-lower-bound (f4617b8)²}

42 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. ↩
No successful run was found on cost-model-r3-lower-bound (d7a3b3a) during the generation of this report, so 0c7981b was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩