codex/cast-bitpacked-pushdown-on-benchmark - Branch - vortex-data/vortex

Add bit-packed widening cast pushdown

#8059Merged

Comparing

codex/cast-bitpacked-pushdown-on-benchmark

(

48a473f

) with

develop

(

1a5079b

)

+72%

Improvement: 7

Regression: 2

Untouched: 1242

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

1251 total

take_filter_list_random_mask_sequential_indices[768, 50]

vortex-array/benches/take_filter.rs

170.2 µs170.3 µs

zstd_compress_u32

vortex/benches/single_encoding_throughput.rs

100.1 ms100.1 ms

chunked_dict_primitive_into_canonical[f32, (1000, 1000, 100)]

vortex-array/benches/chunked_dict_builder.rs

869.2 µs869.4 µs

pushdown_compare[(10000, 4, 8)]

encodings/fsst/benches/fsst_compress.rs

423.6 µs423.7 µs

chunked_dict_primitive_canonical_into[u32, (1000, 10, 10)]

vortex-array/benches/chunked_dict_builder.rs

82.6 µs82.7 µs

alp_decompress_f64

vortex/benches/single_encoding_throughput.rs

1.5 ms1.5 ms

alp_compress_f64

vortex/benches/single_encoding_throughput.rs

1.4 ms1.4 ms

decompress_rd[f32, (100000, 0.1)]

encodings/alp/benches/alp_compress.rs

411.6 µs411.7 µs

decompress_rd[f32, (100000, 0.01)]

encodings/alp/benches/alp_compress.rs

411.6 µs411.7 µs

chunked_dict_primitive_into_canonical[f32, (1000, 10, 100)]

vortex-array/benches/chunked_dict_builder.rs

684.2 µs684.4 µs

take_filter_list_slice_mask_sequential_indices[768, 50]

vortex-array/benches/take_filter.rs

153.9 µs154 µs

take_filter_list_small_random_mask_random_indices[256, 10]

vortex-array/benches/take_filter.rs

171 µs171 µs

chunked_into_canonical[(1000, 100, 16, 16)]

encodings/fsst/benches/fsst_compress.rs

11.4 ms11.4 ms

sequence_compress_u32

vortex/benches/single_encoding_throughput.rs

521.9 µs522 µs

decompress_rd[f32, (10000, 0.0)]

encodings/alp/benches/alp_compress.rs

84.1 µs84.2 µs

dict_compress_u32

vortex/benches/single_encoding_throughput.rs

3.4 ms3.4 ms

compress[(10000, 4)]

encodings/runend/benches/run_end_compress.rs

203.3 µs203.3 µs

compress_fsst[(10000, 4, 8)]

encodings/fsst/benches/fsst_compress.rs

1.6 ms1.6 ms

case_when_nary_10_conditions[1000]

vortex-array/benches/expr/case_when_bench.rs

242.4 µs242.5 µs

chunked_varbinview_opt_into_canonical[(10, 1000)]

vortex-array/benches/chunk_array_builder.rs

2.7 ms2.7 ms

compress_rd[f64, (10000, 0.01)]

encodings/alp/benches/alp_compress.rs

317.6 µs317.7 µs

cuda/bitpacked_patched_u32/unpack/1%[100M]

vortex-cuda/benches/bitpacked_cuda.rs::benches::benchmark_bitunpack_with_patches::cuda

1 ms1 ms

case_when_fragmented[100]

vortex-array/benches/expr/case_when_bench.rs

94.1 µs94.1 µs

list_i32_large

vortex-array/benches/listview_rebuild.rs

344.8 µs344.9 µs

canonicalize_compare[(10000, 16, 4)]

encodings/fsst/benches/fsst_compress.rs

704.4 µs704.6 µs

Commits

Click on a commit to change the comparison range

Base

develop

1a5079b

+77.65%

Add bit-packed widening cast pushdown

468cc1c

15 days ago

by joseph-isaacs

-4.71%

Merge remote-tracking branch 'origin/develop' into codex/cast-bitpacked-pushdown-on-benchmark

6c248bb

15 days ago

by joseph-isaacs

-60.66%

fix

56d3c9e

15 days ago

by joseph-isaacs

+0.01%

fix

3875d53

15 days ago

by joseph-isaacs

+0.26%

fix

cd1b4c2

15 days ago

by joseph-isaacs

+59.3%

fix

48a473f

15 days ago

by joseph-isaacs

Home Terms Privacy Docs