ji/move-take-kernel - Branch - vortex-data/vortex - CodSpeed

Blog Docs Changelog

chore[array]: move take kernel to execute/reduce rules

#6310Merged

Comparing

ji/move-take-kernel

(

0e68b7f

) with

develop

(

e061c55

)

-99%

Improvement: 25

Regression: 26

Untouched: 1087

Skipped: 1265

Benchmarks

2403 total

take_indices[(1000000, 8192)]

encodings/runend/benches/run_end_compress.rs

-99%

27.5 µs1,982 µs

take_indices[(1000000, 4096)]

encodings/runend/benches/run_end_compress.rs

-98%

32.1 µs1,981.9 µs

take_indices[(1000000, 1024)]

encodings/runend/benches/run_end_compress.rs

-97%

59.9 µs2,032.6 µs

decompress[dict_fsst_varbin_string]

vortex/benches/common_encoding_tree_throughput.rs

-93%

16.8 ms226 ms

decompress[dict_fsst_varbin_bp_string]

vortex/benches/common_encoding_tree_throughput.rs

-93%

16.8 ms225.9 ms

take_indices[(1000000, 256)]

encodings/runend/benches/run_end_compress.rs

-92%

171.1 µs2,258.5 µs

take_indices[(100000, 4096)]

encodings/runend/benches/run_end_compress.rs

-91%

22.2 µs235.6 µs

take_indices[(100000, 1024)]

encodings/runend/benches/run_end_compress.rs

-89%

26.6 µs237.8 µs

take_indices[(100000, 256)]

encodings/runend/benches/run_end_compress.rs

-86%

37.7 µs260.6 µs

take_indices[(1000000, 4)]

encodings/runend/benches/run_end_compress.rs

-75%

4 ms16.1 ms

take_indices[(100000, 4)]

encodings/runend/benches/run_end_compress.rs

-74%

425.2 µs1,642.9 µs

take_indices[(10000, 4)]

encodings/runend/benches/run_end_compress.rs

-68%

63.1 µs198.7 µs

take_indices[(1000000, 16)]

encodings/runend/benches/run_end_compress.rs

-66%

2.4 ms7 ms

take_indices[(100000, 16)]

encodings/runend/benches/run_end_compress.rs

-65%

260.1 µs738.7 µs

take_indices[(10000, 256)]

encodings/runend/benches/run_end_compress.rs

-62%

22.7 µs60.2 µs

take_indices[(10000, 1024)]

encodings/runend/benches/run_end_compress.rs

-62%

22.1 µs58.1 µs

take_indices[(10000, 16)]

encodings/runend/benches/run_end_compress.rs

-57%

46.7 µs108.3 µs

take_indices[(1000, 4)]

encodings/runend/benches/run_end_compress.rs

-53%

27.6 µs58.7 µs

take_indices[(1000, 16)]

encodings/runend/benches/run_end_compress.rs

-45%

24.6 µs44.3 µs

take_indices[(1000, 256)]

encodings/runend/benches/run_end_compress.rs

-42%

22.9 µs39.9 µs

take_10_contiguous

encodings/fastlanes/benches/bitpacking_take.rs

-20%

24.9 µs30.9 µs

take_10_stratified

encodings/fastlanes/benches/bitpacking_take.rs

-19%

24 µs29.8 µs

patched_take_10_stratified

encodings/fastlanes/benches/bitpacking_take.rs

-16%

30.9 µs36.9 µs

patched_take_10_contiguous

encodings/fastlanes/benches/bitpacking_take.rs

-14%

31.7 µs37.1 µs

bench_many_nulls[0.9]

vortex-array/benches/dict_unreferenced_mask.rs

-12%

469.4 µs535.2 µs

Commits

Click on a commit to change the comparison range

Base

develop

e061c55

-0.08%

wip

011a071

22 days ago

by joseph-isaacs

-99.69%

wip

44ac66f

22 days ago

by joseph-isaacs

+99.86%

Merge remote-tracking branch 'origin/develop' into ji/move-take-kernel

4281cf3

22 days ago

by joseph-isaacs

-99.86%

wip

53b9d17

22 days ago

by joseph-isaacs

0%

wip

f05ebb9

22 days ago

by joseph-isaacs

0%

wip

5269799

22 days ago

by joseph-isaacs

0%

wip

962bfae

22 days ago

by joseph-isaacs

+1.17%

wip

04e9fea

22 days ago

by joseph-isaacs

0%

wip

f455669

22 days ago

by joseph-isaacs

0%

Merge remote-tracking branch 'origin/develop' into ji/move-take-kernel

b154761

22 days ago

by joseph-isaacs

-0.01%

wip

0e68b7f

22 days ago

by joseph-isaacs

© 2026 CodSpeed Technology

Home Terms Privacy Docs