`RowFn` binary numeric `ScalarFn` rewrite by connortsui20 · Pull Request #9320 · vortex-data/vortex

Performance Regression: -13.72%

⚡ 5 improved benchmarks
❌ 43 regressed benchmarks
✅ 1927 untouched benchmarks
🆕 1 new benchmark
⏩ 89 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

	Mode	Benchmark	`BASE`	`HEAD`	Efficiency
❌	Simulation	`case_when_nary_10_conditions[1000]`	425.9 µs	614.8 µs	-30.72%
❌	Simulation	`baseline_lt[4, 1024]`	98.2 µs	131.1 µs	-25.14%
❌	Simulation	`case_when_nary_equality_lookup[1000]`	323.1 µs	426.2 µs	-24.2%
❌	Simulation	`case_when_nary_10_conditions[10000]`	766.3 µs	1,004.4 µs	-23.71%
❌	Simulation	`baseline_eq[4, 1024]`	97.1 µs	127 µs	-23.53%
❌	Simulation	`baseline_eq[16, 1024]`	98.8 µs	128.5 µs	-23.09%
❌	Simulation	`eq_i64_constant`	372.2 µs	481.3 µs	-22.68%
❌	Simulation	`baseline_lt[16, 1024]`	99.6 µs	128.3 µs	-22.38%
❌	Simulation	`case_when_nary_early_dominant[1000]`	234.4 µs	296.9 µs	-21.06%
❌	Simulation	`case_when_nary_3_conditions[1000]`	236.5 µs	298.7 µs	-20.8%
❌	Simulation	`case_when_nary_equality_lookup[10000]`	526.1 µs	654.3 µs	-19.59%
❌	Simulation	`cold_misaligned[(64, 256)]`	4.4 ms	5.4 ms	-18.59%
❌	Simulation	`case_when_all_false[1000]`	125 µs	152.4 µs	-17.95%
❌	Simulation	`bench_compare_sliced_dict_primitive[(1000, 10000)]`	130.6 µs	158.7 µs	-17.76%
❌	Simulation	`bench_compare_sliced_dict_primitive[(2500, 10000)]`	151.2 µs	182.2 µs	-17.01%
❌	Simulation	`bench_compare_sliced_dict_primitive[(2000, 10000)]`	144.5 µs	173.7 µs	-16.82%
❌	Simulation	`case_when_all_true[1000]`	131.1 µs	156.9 µs	-16.45%
❌	Simulation	`case_when_nary_early_dominant[10000]`	391.7 µs	468.5 µs	-16.39%
❌	Simulation	`case_when_nary_3_conditions[10000]`	394.1 µs	471.1 µs	-16.35%
❌	Simulation	`bench_compare_sliced_dict_primitive[(3333, 10000)]`	165.6 µs	197.1 µs	-15.97%
...	...	...	...	...	...

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 ct/row-fn-numeric (b490007) with ct/row-fn-api (f759998)}

89 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. ↩