Improve visitor performance by reducing `Vec` and String allocations
#6337Comparing
adamg/reduce-allocations-some-cases
(9f62be2
) withdevelop
(133e7dd
)-13%
- Improvement
- 8
- Regression
- 1
- Untouched
- 1129
- Skipped
- 1265
Benchmarks
2403 total
chunked_bool_into_canonical[(1000, 10)]
vortex-array/benches/chunk_array_builder.rs-13%
65.7 µs75.9 µs
chunked_bool_canonical_into[(1000, 10)]
vortex-array/benches/chunk_array_builder.rs+44%
67 µs46.6 µs
chunked_bool_into_canonical[(10, 1000)]
vortex-array/benches/chunk_array_builder.rs+39%
1.9 ms1.3 ms
chunked_opt_bool_into_canonical[(10, 1000)]
vortex-array/benches/chunk_array_builder.rs+36%
2 ms1.4 ms
chunked_bool_into_canonical[(100, 100)]
vortex-array/benches/chunk_array_builder.rs+32%
228.2 µs173.5 µs
chunked_varbinview_into_canonical[(10, 1000)]
vortex-array/benches/chunk_array_builder.rs+20%
3 ms2.5 ms
chunked_opt_bool_into_canonical[(100, 100)]
vortex-array/benches/chunk_array_builder.rs+15%
442.5 µs386.1 µs
patched_take_200k_dispersed
encodings/fastlanes/benches/bitpacking_take.rs+11%
5.6 ms5.1 ms
decompress[u8, (1000, 16)]
encodings/runend/benches/run_end_compress.rs+11%
19.8 µs17.9 µs
chunked_opt_bool_into_canonical[(1000, 10)]
vortex-array/benches/chunk_array_builder.rs+9%
112.7 µs103.5 µs
chunked_dict_primitive_into_canonical[f32, (1000, 100, 10)]
vortex-array/benches/chunked_dict_builder.rs+9%
112.8 µs103.8 µs
chunked_dict_primitive_into_canonical[u32, (1000, 100, 10)]
vortex-array/benches/chunked_dict_builder.rs+8%
111.3 µs102.6 µs
chunked_dict_primitive_into_canonical[f32, (1000, 10, 10)]
vortex-array/benches/chunked_dict_builder.rs+8%
111.2 µs102.6 µs
decompress[u8, (1000, 4)]
encodings/runend/benches/run_end_compress.rs+8%
26.1 µs24.3 µs
chunked_into_canonical[(1000, 50, 8, 4)]
encodings/fsst/benches/fsst_compress.rs+7%
8 ms7.4 ms
chunked_dict_primitive_into_canonical[f32, (1000, 1000, 10)]
vortex-array/benches/chunked_dict_builder.rs+7%
129.2 µs120.4 µs
compress[(1000, 256)]
encodings/runend/benches/run_end_compress.rs+7%
52.9 µs49.4 µs
chunked_dict_primitive_into_canonical[u32, (1000, 1000, 10)]
vortex-array/benches/chunked_dict_builder.rs+7%
128 µs119.5 µs
chunked_dict_primitive_into_canonical[u32, (1000, 100, 100)]
vortex-array/benches/chunked_dict_builder.rs+7%
838.2 µs784.1 µs
chunked_dict_primitive_into_canonical[u32, (1000, 10, 100)]
vortex-array/benches/chunked_dict_builder.rs+7%
821.2 µs768.8 µs
true_count_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+7%
917.8 ns859.4 ns
chunked_dict_primitive_into_canonical[u64, (1000, 100, 10)]
vortex-array/benches/chunked_dict_builder.rs+7%
155.6 µs145.9 µs
chunked_varbinview_opt_into_canonical[(10, 1000)]
vortex-array/benches/chunk_array_builder.rs+7%
7.3 ms6.8 ms
chunked_into_canonical[(1000, 50, 8, 16)]
encodings/fsst/benches/fsst_compress.rs+6%
9.1 ms8.6 ms
true_count_arrow_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+6%
980.3 ns921.9 ns
Commits
Click on a commit to change the comparison rangeBase
develop
133e7dd
+36.4%
why not
015b5d7
3 days ago
by AdamGS-0.01%
Revert filter change
8172fcb
3 days ago
by AdamGS-49.77%
nbuffers, nchildren, more stuff
9f62be2
3 hours ago
by AdamGS