claude/confident-hamilton-mZIEo - Branch - vortex-data/vortex

perf(fastlanes): fuse bit-packed compare into a transposed mask + untranspose

#8239

Comparing

claude/confident-hamilton-mZIEo

(

0649f03

) with

develop

(

fd36910

)

+59%

Improvement: 236

Regression: 1

Untouched: 1270

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.

Benchmarks

1507 total

compress_alp[f64, (1000, 0.01, 1.0)]

encodings/alp/benches/alp_compress.rs

258.9 µs259.2 µs

compress_alp[f64, (1000, 0.1, 1.0)]

encodings/alp/benches/alp_compress.rs

260.5 µs260.8 µs

take_struct_sequential_indices

vortex-array/benches/take_struct.rs

48.9 µs49 µs

pack_return_dtype[500]

vortex-array/benches/expr/large_struct_pack.rs

318.4 µs318.8 µs

bench_compare_sliced_dict_primitive[(10000, 10000)]

vortex-array/benches/dict_compare.rs

160.3 µs160.5 µs

baseline_lt[16, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

216.1 µs216.4 µs

bench_compare_sliced_dict_varbinview[(2000, 10000)]

vortex-array/benches/dict_compare.rs

140.5 µs140.7 µs

take_10k_first_chunk_only

encodings/fastlanes/benches/bitpacking_take.rs

271.2 µs271.6 µs

take_10k_dispersed

encodings/fastlanes/benches/bitpacking_take.rs

284.9 µs285.3 µs

take_map[(0.01, 0.5)]

vortex-array/benches/take_patches.rs

600.4 µs601.3 µs

compress_alp[f64, (1000, 0.0, 0.25)]

encodings/alp/benches/alp_compress.rs

222.3 µs222.6 µs

compress_alp[f64, (1000, 0.0, 0.95)]

encodings/alp/benches/alp_compress.rs

222.2 µs222.6 µs

push_n_vortex_buffer[u16, 16384]

vortex-buffer/benches/vortex_buffer.rs

18.5 µs18.5 µs

baseline_eq[16, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

228.4 µs228.8 µs

baseline_lt[4, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

199.4 µs199.7 µs

append_arrow_buffer[2048]

vortex-buffer/benches/vortex_bitbuffer.rs

17.3 µs17.3 µs

patched_take_10k_contiguous_not_patches

encodings/fastlanes/benches/bitpacking_take.rs

290 µs290.5 µs

bench_compare_sliced_dict_varbinview[(9999, 10000)]

vortex-array/benches/dict_compare.rs

254.4 µs254.8 µs

take_10k_random

encodings/fastlanes/benches/bitpacking_take.rs

237.9 µs238.4 µs

compress_alp[f64, (1000, 0.1, 0.95)]

encodings/alp/benches/alp_compress.rs

274.3 µs274.8 µs

take_10k_contiguous

encodings/fastlanes/benches/bitpacking_take.rs

290.6 µs291.2 µs

compress_alp[f64, (1000, 0.1, 0.25)]

encodings/alp/benches/alp_compress.rs

269.7 µs270.2 µs

bench_dict_mask[(0.5, 0.1)]

vortex-array/benches/dict_mask.rs

1.9 ms1.9 ms

baseline_eq[4, 65536]

encodings/fastlanes/benches/bitpack_compare.rs

183.5 µs183.9 µs

compress_alp[f64, (1000, 0.01, 0.25)]

encodings/alp/benches/alp_compress.rs

266.1 µs266.7 µs

Commits

Click on a commit to change the comparison range

Base

develop

fd36910

+46%

perf(fastlanes): fuse bit-packed compare into a transposed mask + untranspose

48da899

22 hours ago

by claude

+17.53%

Merge branch 'develop' into claude/confident-hamilton-mZIEo

ab9c8d6

5 hours ago

by joseph-isaacs

-4.72%

ci(wasm): pin in-development FastLanes in the excluded wasm-test workspace

08ed4a4

4 hours ago

by claude

+5.42%

wip

816032b

2 hours ago

by joseph-isaacs

-0.47%

wip

e4dd660

2 hours ago

by joseph-isaacs

-65.08%

wip

211903c

2 hours ago

by joseph-isaacs

+60.04%

wip

0649f03

20 minutes ago

by joseph-isaacs

Home Terms Privacy Docs