Comparing
ji/take-inline-2
(eda5eb4
) withdevelop
(60fa538
)-25%
- Improvement
- 8
- Regression
- 2
- Untouched
- 1128
- Skipped
- 1265
Benchmarks
2403 total
chunked_bool_into_canonical[(1000, 10)]
vortex-array/benches/chunk_array_builder.rs-25%
66 µs88.6 µs
take_indices[(1000000, 8192)]
encodings/runend/benches/run_end_compress.rs-13%
7.5 µs8.6 µs
take_indices[(10000, 4)]
encodings/runend/benches/run_end_compress.rs+31%
9.1 µs7 µs
take_struct_wide[8]
vortex-array/benches/take_struct.rs+21%
9.2 µs7.6 µs
take_indices[(1000, 16)]
encodings/runend/benches/run_end_compress.rs+20%
8.4 µs7 µs
take_indices[(1000, 256)]
encodings/runend/benches/run_end_compress.rs+19%
8.3 µs7 µs
take_struct_simple
vortex-array/benches/take_struct.rs+16%
7.8 µs6.7 µs
take_indices[(10000, 16)]
encodings/runend/benches/run_end_compress.rs+15%
8 µs7 µs
take_struct_sequential_indices
vortex-array/benches/take_struct.rs+13%
8.8 µs7.8 µs
take_10_contiguous
encodings/fastlanes/benches/bitpacking_take.rs+12%
7.6 µs6.8 µs
patched_take_10_contiguous
encodings/fastlanes/benches/bitpacking_take.rs+10%
9.4 µs8.6 µs
take_indices[(100000, 4)]
encodings/runend/benches/run_end_compress.rs+10%
9.1 µs8.3 µs
take_fsl_nullable_random[16, 100]
vortex-array/benches/take_fsl.rs+9%
9 µs8.3 µs
take_fsl_random[1024, 1000]
vortex-array/benches/take_fsl.rs+8%
7.3 µs6.8 µs
take_fsl_random[4096, 1000]
vortex-array/benches/take_fsl.rs+8%
7.3 µs6.8 µs
take_fsl_random[256, 1000]
vortex-array/benches/take_fsl.rs+8%
7.3 µs6.8 µs
take_fsl_random[16, 1000]
vortex-array/benches/take_fsl.rs+8%
7.3 µs6.8 µs
take_fsl_random[4096, 100]
vortex-array/benches/take_fsl.rs+8%
7.3 µs6.8 µs
take_fsl_random[64, 1000]
vortex-array/benches/take_fsl.rs+7%
7.3 µs6.8 µs
take_fsl_random[1024, 100]
vortex-array/benches/take_fsl.rs+7%
7.3 µs6.8 µs
take_fsl_random[64, 100]
vortex-array/benches/take_fsl.rs+7%
7.3 µs6.8 µs
take_fsl_random[16, 100]
vortex-array/benches/take_fsl.rs+7%
7.3 µs6.8 µs
take_fsl_random[256, 100]
vortex-array/benches/take_fsl.rs+7%
7.3 µs6.8 µs
take_indices[(100000, 16)]
encodings/runend/benches/run_end_compress.rs+7%
7.5 µs7 µs
compress[(1000, 256)]
encodings/runend/benches/run_end_compress.rs+7%
52.8 µs49.4 µs
Commits
Click on a commit to change the comparison rangeBase
develop
60fa538
+0.03%
chore[array]: inline take compute function p2
b4edc1a
12 hours ago
by joseph-isaacs+0.01%
chore[array]: inline take compute function p2
03e4ac6
12 hours ago
by joseph-isaacs-25.55%
chore[array]: inline take compute function p2
eda5eb4
12 hours ago
by joseph-isaacs