claude/nice-archimedes-yjGyO - Branch - vortex-data/vortex

feat(vortex-row): row-oriented byte encoder (size + encode passes)

#8253

Comparing

claude/nice-archimedes-yjGyO

(

946e18c

) with

develop

(

e484daf

)

-13%

Improvement: 1

Regression: 1

Untouched: 1505

New: 10

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

1517 total

chunked_bool_canonical_into[(1000, 10)]

vortex-array/benches/chunk_array_builder.rs

-32%

31.7 µs46.6 µs

bitwise_not_vortex_buffer_mut[128]

vortex-buffer/benches/vortex_bitbuffer.rs

+12%

275.3 ns246.1 ns

struct_mixed_vortex

vortex-row/benches/row_encode.rs

N/A

N/A22.9 ms

primitive_i64_vortex

vortex-row/benches/row_encode.rs

N/A

N/A1.5 ms

fsst_decompress_only

vortex-row/benches/fsst_row_encode.rs

N/A

N/A16.4 ms

plain_row_encode_only

vortex-row/benches/fsst_row_encode.rs

N/A

N/A14.9 ms

primitive_i64_arrow_row

vortex-row/benches/row_encode.rs

N/A

N/A2.4 ms

fsst_fast_fused

vortex-row/benches/fsst_row_encode.rs

N/A

N/A29.5 ms

fsst_unpack_then_convert

vortex-row/benches/fsst_row_encode.rs

N/A

N/A30.6 ms

struct_mixed_arrow_row

vortex-row/benches/row_encode.rs

N/A

N/A18.7 ms

utf8_arrow_row

vortex-row/benches/row_encode.rs

N/A

N/A8.6 ms

utf8_vortex

vortex-row/benches/row_encode.rs

N/A

N/A9.2 ms

bitwise_not_vortex_buffer_mut[1024]

vortex-buffer/benches/vortex_bitbuffer.rs

+9%

336.9 ns307.8 ns

bitwise_not_vortex_buffer_mut[2048]

vortex-buffer/benches/vortex_bitbuffer.rs

+8%

400.6 ns371.4 ns

true_count_arrow_buffer[128]

vortex-buffer/benches/vortex_bitbuffer.rs

+8%

813.9 ns755.6 ns

cuda/dict_nowiden_u32_to_u32/dispatch_u32[100M]

vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::cuda

+6%

1,015.8 µs961.5 µs

cuda/dict_widen_u16_to_u32/dispatch_u32[100M]

vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::cuda

+6%

1,015.8 µs962.6 µs

cuda/alp_f32/0%[100M]

vortex-cuda/benches/alp_cuda.rs::benches::benchmark_alp_decode::cuda

+5%

2.5 ms2.4 ms

cuda/for_bitpacked_6bw/dispatch_u32[100M]

vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::cuda

+5%

1.1 ms1 ms

bitwise_and_arrow_buffer[128]

vortex-buffer/benches/vortex_bitbuffer.rs

+3%

1.8 µs1.8 µs

bitwise_or_arrow_buffer[128]

vortex-buffer/benches/vortex_bitbuffer.rs

+3%

1.9 µs1.8 µs

bitwise_and_arrow_buffer[1024]

vortex-buffer/benches/vortex_bitbuffer.rs

+3%

2 µs2 µs

true_count_arrow_buffer[1024]

vortex-buffer/benches/vortex_bitbuffer.rs

+3%

1,023.9 ns994.7 ns

bitwise_or_arrow_buffer[1024]

vortex-buffer/benches/vortex_bitbuffer.rs

+3%

2.1 µs2 µs

true_count_arrow_buffer[2048]

vortex-buffer/benches/vortex_bitbuffer.rs

+3%

1.1 µs1.1 µs

Commits

Click on a commit to change the comparison range

Base

develop

e484daf

+30.19%

feat(vortex-row): add row-oriented byte encoder crate

4a30a9c

14 hours ago

by joseph-isaacs

-30.19%

ci(vortex-row): run row_encode benchmarks on CodSpeed

083c7f3

13 hours ago

by joseph-isaacs

+28.93%

vortex-row: walk VarBinView rows directly in the encoder hot loop

ee049ae

11 hours ago

by joseph-isaacs

-7.76%

vortex-row: satisfy cast-truncation lints in ported hot paths

65a24f9

11 hours ago

by joseph-isaacs

-53.1%

vortex-row: arithmetic-write fast path for fixed-before-varlen columns

2711504

9 hours ago

by joseph-isaacs

vortex-row: benchmark FSST row-encoding (unpack-then-convert vs phases)

a287003

6 hours ago

by joseph-isaacs

vortex-row: rustfmt the fsst row-encode benchmark

4f155e2

6 hours ago

by joseph-isaacs

+19.23%

vortex-row: add fused fast-path FSST row-encode benchmark

946e18c

6 hours ago

by joseph-isaacs

Home Terms Privacy Docs