ad/patches-no-search-sorted - Branch - vortex-data/vortex - CodSpeed

Blog Docs Changelog

feat: constant time patches slicing

#4913Merged

Comparing

ad/patches-no-search-sorted

(

78ca82a

) with

develop

(

370ad2b

)

-22%

Improvement: 6

Regression: 4

Untouched: 596

Archived: 566

Benchmarks

1172 total

into_canonical_nullable[(10000, 100, 0.1)]

encodings/fastlanes/benches/canonicalize_bench.rs

-22%

3.3 ms4.2 ms

into_canonical_nullable[(10000, 10, 0.1)]

encodings/fastlanes/benches/canonicalize_bench.rs

-18%

357.6 µs438.7 µs

into_canonical_nullable[(10000, 1, 0.1)]

encodings/fastlanes/benches/canonicalize_bench.rs

-17%

51.6 µs62.5 µs

into_canonical_nullable[(10000, 1, 0.0)]

encodings/fastlanes/benches/canonicalize_bench.rs

-16%

42 µs50 µs

canonical_into_non_nullable[(10000, 10, 0.0)]

encodings/fastlanes/benches/canonicalize_bench.rs

+45%

277.8 µs191.4 µs

canonical_into_non_nullable[(10000, 10, 0.01)]

encodings/fastlanes/benches/canonicalize_bench.rs

+41%

297 µs210.8 µs

canonical_into_non_nullable[(10000, 10, 0.1)]

encodings/fastlanes/benches/canonicalize_bench.rs

+28%

393.7 µs307.1 µs

take_200k_first_chunk_only

encodings/fastlanes/benches/bitpacking_take.rs

+26%

4.3 ms3.4 ms

take_200k_dispersed

encodings/fastlanes/benches/bitpacking_take.rs

+24%

4.6 ms3.7 ms

patched_take_200k_first_chunk_only

encodings/fastlanes/benches/bitpacking_take.rs

+12%

5.4 ms4.8 ms

patched_take_200k_dispersed

encodings/fastlanes/benches/bitpacking_take.rs

+8%

5.1 ms4.7 ms

map_each[BufferMut<i32>, 128]

vortex-buffer/benches/vortex_buffer.rs

+7%

465.8 ns436.7 ns

canonical_into_non_nullable[(10000, 1, 0.01)]

encodings/fastlanes/benches/canonicalize_bench.rs

+6%

40.8 µs38.4 µs

canonical_into_non_nullable[(10000, 1, 0.0)]

encodings/fastlanes/benches/canonicalize_bench.rs

+6%

37.1 µs34.9 µs

old_bp_prim_test_between[i32, 2048]

encodings/fastlanes/benches/compute_between.rs::bitpack

+5%

93.1 µs88.8 µs

take_search[(0.005, 0.05)]

vortex-array/benches/take_patches.rs

+4%

53.6 µs51.4 µs

canonical_into_non_nullable[(10000, 1, 0.1)]

encodings/fastlanes/benches/canonicalize_bench.rs

+4%

50.1 µs48.1 µs

into_canonical_non_nullable[(10000, 10, 0.0)]

encodings/fastlanes/benches/canonicalize_bench.rs

+4%

279.7 µs269.8 µs

into_canonical_non_nullable[(10000, 10, 0.01)]

encodings/fastlanes/benches/canonicalize_bench.rs

+3%

299.1 µs289.1 µs

into_canonical_nullable[(10000, 100, 0.0)]

encodings/fastlanes/benches/canonicalize_bench.rs

+3%

3.2 ms3.1 ms

decompress[u8, (1000, 256)]

encodings/runend/benches/run_end_compress.rs

+3%

9.1 µs8.9 µs

decompress[u8, (1000, 16)]

encodings/runend/benches/run_end_compress.rs

+2%

10.1 µs9.9 µs

take_map[(0.005, 0.05)]

vortex-array/benches/take_patches.rs

+2%

46.4 µs45.4 µs

chunked_bool_into_canonical[(1000, 10)]

vortex-array/benches/chunk_array_builder.rs

+2%

17.7 µs17.4 µs

new_raw_prim_test_between[f32, 2048]

encodings/fastlanes/benches/compute_between.rs::primitive

+2%

25.6 µs25.1 µs

Commits

Click on a commit to change the comparison range

Base

develop

370ad2b

-22.69%

fix bitpacking build

5076f34

4 months ago

by 0ax1

0%

typo

70b3bee

4 months ago

by 0ax1

0%

simplify bitpacking chunk offset recording

7d0f2b0

4 months ago

by 0ax1

0%

option

6ce95c2

4 months ago

by 0ax1

+0.26%

further simplify patch index collection for bitpacking

b6ac04c

4 months ago

by 0ax1

+0.19%

clarify names

33ef491

4 months ago

by 0ax1

0%

chunk_offset_at helper fn

280b9c7

4 months ago

by 0ax1

+0.09%

add comment for bitpacking layout

174ecc0

4 months ago

by 0ax1

-0.08%

as_ptype

4358b01

4 months ago

by 0ax1

+0.1%

validity idx

78ca82a

4 months ago

by 0ax1

© 2026 CodSpeed Technology

Home Terms Privacy Docs