claude/optimize-bitpack-comparison-KGPS3 - Branch - vortex-data/vortex

Fast-path comparison and constant encoding for bit-packed arrays

#8011Closed

Comparing

claude/optimize-bitpack-comparison-KGPS3

(

a20f09d

) with

develop

(

52e26d1

)

+4%

Improvement: 2

Regression: 1

Untouched: 1218

New: 24

Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data. For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

No successful run was found on develop ( faf7e42) during the generation of this report, so 52e26d1 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report.

Benchmarks

1245 total

chunked_varbinview_opt_canonical_into[(1000, 10)]

vortex-array/benches/chunk_array_builder.rs

-16%

188 µs224.7 µs

chunked_varbinview_into_canonical[(1000, 10)]

vortex-array/benches/chunk_array_builder.rs

+20%

211.9 µs176.3 µs

chunked_varbinview_into_canonical[(100, 100)]

vortex-array/benches/chunk_array_builder.rs

+11%

358.4 µs323.3 µs

fast_encode[4, 1024]

encodings/fastlanes/benches/bitpack_constant.rs

N/A

N/A11.5 µs

full_encode[16, 65536]

encodings/fastlanes/benches/bitpack_constant.rs

N/A

N/A358 µs

full_encode[16, 1024]

encodings/fastlanes/benches/bitpack_constant.rs

N/A

N/A17.3 µs

fast_encode[4, 65536]

encodings/fastlanes/benches/bitpack_constant.rs

N/A

N/A30.6 µs

full_encode[4, 1024]

encodings/fastlanes/benches/bitpack_constant.rs

N/A

N/A19.2 µs

baseline_eq[4, 1024]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A64 µs

baseline_lt[4, 1024]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A79 µs

fast_lt_out_of_range[4, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A35 µs

fast_encode[16, 65536]

encodings/fastlanes/benches/bitpack_constant.rs

N/A

N/A81.5 µs

full_encode[4, 65536]

encodings/fastlanes/benches/bitpack_constant.rs

N/A

N/A313.6 µs

baseline_eq[16, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A288.1 µs

baseline_lt[16, 1024]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A65.1 µs

baseline_eq[16, 1024]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A64.6 µs

baseline_eq[4, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A243.2 µs

fast_lt_out_of_range[16, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A35.1 µs

fast_eq_out_of_range[16, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A35.6 µs

baseline_lt[16, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A275.8 µs

baseline_lt[4, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A259 µs

fast_eq_out_of_range[16, 1024]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A26.8 µs

fast_eq_out_of_range[4, 1024]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A26.8 µs

fast_encode[16, 1024]

encodings/fastlanes/benches/bitpack_constant.rs

N/A

N/A15.5 µs

fast_lt_out_of_range[16, 1024]

encodings/fastlanes/benches/bitpack_compare.rs

N/A

N/A25.6 µs

Commits

Click on a commit to change the comparison range

Base

develop

52e26d1

+3.69%

Extend bit-packed compare-constant to ordering, add bitpack_constant kernel and benches, plan in-range ordering

a20f09d

5 days ago

by claude

Home Terms Privacy Docs