Decimal sum uses smallest physical type that supports output dtype
#7008Comparing
rk/decimalbase
(06eaf8e
) withdevelop
(61bc044
)+11%
- Improvement
- 1
- Untouched
- 1008
- Skipped
- 1515
Benchmarks
2524 total
binary_search_std
vortex-array/benches/search_sorted.rs+11%
582.8 ns524.4 ns
chunked_opt_bool_canonical_into[(100, 100)]
vortex-array/benches/chunk_array_builder.rs+8%
430.2 µs397.2 µs
binary_search_vortex
vortex-array/benches/search_sorted.rs+8%
807.2 ns748.9 ns
set_indices_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+5%
1.2 µs1.2 µs
chunked_bool_canonical_into[(10, 1000)]
vortex-array/benches/chunk_array_builder.rs+3%
1.7 ms1.7 ms
chunked_bool_canonical_into[(100, 100)]
vortex-array/benches/chunk_array_builder.rs+2%
204.5 µs199.8 µs
chunked_opt_bool_canonical_into[(10, 1000)]
vortex-array/benches/chunk_array_builder.rs+2%
2.2 ms2.2 ms
chunked_opt_bool_into_canonical[(10, 1000)]
vortex-array/benches/chunk_array_builder.rs+2%
2.5 ms2.4 ms
take_map[(0.005, 1.0)]
vortex-array/benches/take_patches.rs+2%
2.4 ms2.4 ms
patched_take_10_contiguous
encodings/fastlanes/benches/bitpacking_take.rs+2%
36.9 µs36.2 µs
chunked_opt_bool_into_canonical[(100, 100)]
vortex-array/benches/chunk_array_builder.rs+2%
438.3 µs430.4 µs
transpose_scalar
encodings/fastlanes/benches/bit_transpose.rs+2%
1.8 µs1.7 µs
null_count_run_end[(10000, 1024, 0.1)]
encodings/runend/benches/run_end_null_count.rs+1%
35.3 µs34.9 µs
compress[(1000, 256)]
encodings/runend/benches/run_end_compress.rs+1%
59.8 µs59 µs
null_count_run_end[(10000, 256, 0.5)]
encodings/runend/benches/run_end_null_count.rs+1%
36.9 µs36.5 µs
append_buffer_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
4.4 µs4.4 µs
bitwise_not_vortex_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
4.4 µs4.4 µs
compress[(1000, 16)]
encodings/runend/benches/run_end_compress.rs+1%
63.4 µs62.6 µs
take_map[(0.01, 0.1)]
vortex-array/benches/take_patches.rs+1%
321.1 µs317 µs
bitwise_not_vortex_buffer[2048]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
4.6 µs4.6 µs
i32_small_overlapping
vortex-array/benches/listview_rebuild.rs+1%
69.1 µs68.2 µs
null_count_run_end[(10000, 1024, 0.5)]
encodings/runend/benches/run_end_null_count.rs+1%
35.3 µs34.9 µs
compress_alp[f32, (1000, 0.01, 1.0)]
encodings/alp/benches/alp_compress.rs+1%
155.5 µs153.6 µs
compress_alp[f32, (1000, 0.1, 1.0)]
encodings/alp/benches/alp_compress.rs+1%
157.3 µs155.5 µs
compress_alp[f32, (1000, 0.01, 0.25)]
encodings/alp/benches/alp_compress.rs+1%
164.9 µs163.1 µs
Commits
Click on a commit to change the comparison rangeBase
develop
61bc044
-0.06%
Decimal sum uses smallest physical type that supports output dtype
d2c3045
22 hours ago
by robert3005-15.45%
borrow
03b33e6
17 hours ago
by robert3005+15.53%
lesscode
8949240
17 hours ago
by robert3005-0.02%
fixes
8658ef1
17 hours ago
by robert3005+11.12%
Update vortex-array/src/aggregate_fn/fns/sum/decimal.rs
dbfc61e
6 hours ago
by robert30050%
format
06eaf8e
6 hours ago
by robert3005