Comparing
ji/lazy-invert
(9e44343
) withdevelop
(6e1e9d5
)+11%
- Improvement
- 1
- Untouched
- 1134
- Skipped
- 1268
Benchmarks
2403 total
patched_take_200k_dispersed
encodings/fastlanes/benches/bitpacking_take.rs+11%
5.6 ms5.1 ms
binary_search_vortex
vortex-array/benches/search_sorted.rs+7%
938.9 ns880.6 ns
set_indices_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+5%
1.2 µs1.1 µs
new_bp_prim_test_between[i32, 2048]
encodings/fastlanes/benches/compute_between.rs::bitpack+5%
42.4 µs40.3 µs
binary_search_std
vortex-array/benches/search_sorted.rs+4%
741.4 ns712.2 ns
new_alp_prim_test_between[f32, 2048]
encodings/fastlanes/benches/compute_between.rs::alp+4%
53.2 µs51.3 µs
decompress_rd[f32, 10000]
encodings/alp/benches/alp_compress.rs+3%
265 µs256.8 µs
true_count_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
1,005.3 ns976.1 ns
chunked_dict_primitive_into_canonical[u64, (1000, 100, 100)]
vortex-array/benches/chunked_dict_builder.rs+2%
1.2 ms1.2 ms
chunked_dict_primitive_into_canonical[f32, (1000, 1000, 100)]
vortex-array/benches/chunked_dict_builder.rs+2%
967.3 µs944.8 µs
chunked_dict_primitive_into_canonical[u64, (1000, 10, 10)]
vortex-array/benches/chunked_dict_builder.rs+2%
146.8 µs143.5 µs
chunked_dict_primitive_canonical_into[f32, (1000, 10, 100)]
vortex-array/benches/chunked_dict_builder.rs+2%
771.4 µs754.4 µs
new_alp_prim_test_between[f64, 2048]
encodings/fastlanes/benches/compute_between.rs::alp+2%
54.5 µs53.5 µs
decompress[u16, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+2%
18.8 µs18.4 µs
chunked_dict_primitive_into_canonical[f64, (1000, 100, 10)]
vortex-array/benches/chunked_dict_builder.rs+2%
149.6 µs147.1 µs
decompress[u8, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+2%
16.5 µs16.2 µs
chunked_dict_primitive_into_canonical[f64, (1000, 1000, 100)]
vortex-array/benches/chunked_dict_builder.rs+2%
1.6 ms1.5 ms
chunked_dict_primitive_canonical_into[f64, (1000, 10, 100)]
vortex-array/benches/chunked_dict_builder.rs+2%
1.2 ms1.1 ms
decompress[u8, (1000, 16)]
encodings/runend/benches/run_end_compress.rs+1%
19.1 µs18.8 µs
take_map[(0.05, 0.05)]
vortex-array/benches/take_patches.rs+1%
464 µs457.3 µs
decompress[u32, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+1%
19.4 µs19.1 µs
new_bp_prim_test_between[i64, 2048]
encodings/fastlanes/benches/compute_between.rs::bitpack+1%
46.7 µs46.1 µs
decompress[u64, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+1%
20.8 µs20.5 µs
patched_take_10_stratified
encodings/fastlanes/benches/bitpacking_take.rs+1%
37.7 µs37.2 µs
append_buffer_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
4.4 µs4.4 µs
Commits
Click on a commit to change the comparison rangeBase
develop
6e1e9d5
+11.33%
feat[array]: refactor invert compute kernel to lazy `not` via ScalarFnArray
0f19499
14 days ago
by joseph-isaacs0%
feat[array]: refactor invert compute kernel to lazy `not` via ScalarFnArray
cf47bc6
14 days ago
by joseph-isaacs-0.01%
Merge branch 'develop' into ji/lazy-invert
9e44343
14 days ago
by joseph-isaacs