perf: aggregate min/max
#8061Comparing
claude/great-edison-jrGY0
(6a44f51
) withdevelop
(495f30e
)+14%
- Improvement
- 10
- Regression
- 1
- Untouched
- 1240
- New
- 10
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.Different runtime environments detected
Some benchmarks with significant performance changes were compared across different runtime environments, which may affect the accuracy of the results. Learn more about regression causes.Benchmarks
1261 total
chunked_varbinview_opt_canonical_into[(1000, 10)]
vortex-array/benches/chunk_array_builder.rs-17%
187.5 µs224.7 µs
for_compress_i32
vortex/benches/single_encoding_throughput.rs+70%
753.2 µs444.1 µs
chunked_varbinview_into_canonical[(1000, 10)]
vortex-array/benches/chunk_array_builder.rs+20%
211.8 µs176.2 µs
cuda/bitpacked_u8/unpack/3bw[100M]
vortex-cuda/benches/bitpacked_cuda.rs::benches::benchmark_bitunpack::cuda+18%
353 µs299.9 µs
encode_primitives[u8, (10000, 8)]
vortex-array/benches/dict_compress.rs+13%
315.2 µs279.2 µs
encode_primitives[u8, (10000, 4)]
vortex-array/benches/dict_compress.rs+13%
314.2 µs278.4 µs
encode_primitives[u8, (10000, 2)]
vortex-array/benches/dict_compress.rs+13%
313.9 µs278.2 µs
encode_primitives[u8, (10000, 32)]
vortex-array/benches/dict_compress.rs+13%
317.7 µs282.4 µs
encode_primitives[u8, (10000, 512)]
vortex-array/benches/dict_compress.rs+12%
334.9 µs299.1 µs
chunked_varbinview_into_canonical[(100, 100)]
vortex-array/benches/chunk_array_builder.rs+11%
358.3 µs323.3 µs
take_10k_contiguous
encodings/fastlanes/benches/bitpacking_take.rs+10%
308.6 µs280.1 µs
max_i32
vortex-array/benches/aggregate_max.rsN/A
N/A222.6 µs
max_i32_nulls_clustered
vortex-array/benches/aggregate_max.rsN/A
N/A254.3 µs
max_i32_nulls_scattered
vortex-array/benches/aggregate_max.rsN/A
N/A1.6 ms
max_i64
vortex-array/benches/aggregate_max.rsN/A
N/A436.3 µs
sum_i32
vortex-array/benches/aggregate_sum.rsN/A
N/A222.1 µs
sum_i32_nulls_clustered
vortex-array/benches/aggregate_sum.rsN/A
N/A236.2 µs
sum_i32_nulls_scattered
vortex-array/benches/aggregate_sum.rsN/A
N/A1.6 ms
sum_i64
vortex-array/benches/aggregate_sum.rsN/A
N/A600.4 µs
sum_u32
vortex-array/benches/aggregate_sum.rsN/A
N/A222.1 µs
max_f64
vortex-array/benches/aggregate_max.rsN/A
N/A1.1 ms
dict_compress_u32
vortex/benches/single_encoding_throughput.rs+10%
3.4 ms3.1 ms
cast_execute[(262144, 1, 0.0)]
encodings/fastlanes/benches/cast_bitpacked.rs+9%
712 µs652.3 µs
new_alp_prim_test_between[f32, 32768]
encodings/fastlanes/benches/compute_between.rs::alp+9%
167.1 µs153.3 µs
cast_execute[(262144, 1, 0.01)]
encodings/fastlanes/benches/cast_bitpacked.rs+9%
741.7 µs681.8 µs
Commits
Click on a commit to change the comparison rangeBase
develop
495f30e
+28.11%
Vectorize integer min/max for all-valid arrays
e9a31ad
1 day ago
by claude-28.26%
Simplify aggregate max benchmark to one bench per type
ac47a14
1 day ago
by claude+32.2%
Reduce aggregate max benchmark array length to 100k
08abd6a
1 day ago
by claude-17.21%
Vectorize integer sum via chunked widening accumulator
8b98b5d
21 hours ago
by claude-0.35%
Vectorize integer sum and min/max over nullable arrays
6a44f51
3 hours ago
by claude