Comparing
claude/benchmarks-v3-emitter-split
(5bd762b
) withdevelop
(e0a0527
)-17%
- Improvement
- 10
- Regression
- 4
- Untouched
- 1155
- Skipped
- 138
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.Benchmarks
1307 total
decompress_rd[f64, (100000, 0.1)]
encodings/alp/benches/alp_compress.rs-17%
842.5 µs1,020.9 µs
decompress_rd[f64, (100000, 0.01)]
encodings/alp/benches/alp_compress.rs-17%
842.6 µs1,020.7 µs
decompress_rd[f32, (100000, 0.1)]
encodings/alp/benches/alp_compress.rs-15%
495 µs582.7 µs
decompress_rd[f32, (100000, 0.01)]
encodings/alp/benches/alp_compress.rs-15%
495 µs582.7 µs
take_10k_first_chunk_only
encodings/fastlanes/benches/bitpacking_take.rs+20%
270.8 µs226 µs
take_10k_dispersed
encodings/fastlanes/benches/bitpacking_take.rs+19%
284.8 µs239.8 µs
decompress_rd[f32, (100000, 0.0)]
encodings/alp/benches/alp_compress.rs+18%
583.4 µs495.8 µs
decompress_rd[f64, (10000, 0.01)]
encodings/alp/benches/alp_compress.rs+14%
138.6 µs122.1 µs
decompress_rd[f64, (10000, 0.1)]
encodings/alp/benches/alp_compress.rs+13%
138.7 µs122.2 µs
decompress_rd[f64, (10000, 0.0)]
encodings/alp/benches/alp_compress.rs+13%
138.5 µs122.2 µs
patched_take_10k_adversarial
encodings/fastlanes/benches/bitpacking_take.rs+13%
259 µs228.7 µs
patched_take_10k_first_chunk_only
encodings/fastlanes/benches/bitpacking_take.rs+11%
302.4 µs272.1 µs
patched_take_10k_dispersed
encodings/fastlanes/benches/bitpacking_take.rs+11%
316 µs285.6 µs
decompress_rd[f32, (10000, 0.0)]
encodings/alp/benches/alp_compress.rs+10%
94.5 µs85.9 µs
decompress_rd[f32, (10000, 0.1)]
encodings/alp/benches/alp_compress.rs+10%
90.1 µs82 µs
decompress_rd[f32, (10000, 0.01)]
encodings/alp/benches/alp_compress.rs+10%
90 µs82 µs
dict_compress_string
vortex/benches/single_encoding_throughput.rs+7%
8.7 ms8.2 ms
cuda/runend_100runs/dispatch_u32[100M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::cuda+6%
891.9 µs838.7 µs
cuda/dict_widen_u8_to_u32/dispatch_u32[100M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::cuda+5%
1,005.6 µs954.4 µs
cuda/for_bitpacked_6bw/dispatch_u32[100M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::cuda+5%
1.1 ms1 ms
compress_rd[f64, (100000, 0.0)]
encodings/alp/benches/alp_compress.rs+5%
2.9 ms2.8 ms
compress_rd[f64, (100000, 0.01)]
encodings/alp/benches/alp_compress.rs+5%
2.9 ms2.8 ms
compress_rd[f64, (100000, 0.1)]
encodings/alp/benches/alp_compress.rs+4%
2.9 ms2.8 ms
decompress_rd[f64, (100000, 0.0)]
encodings/alp/benches/alp_compress.rs+4%
1,021.4 µs979 µs
take_map[(0.1, 0.5)]
vortex-array/benches/take_patches.rs+3%
1,007.2 µs976.1 µs
Commits
Click on a commit to change the comparison rangeBase
develop
e0a0527
-24.93%
feat(bench): emit v3 JSONL records and dual-write to bench server
35d59ae
6 hours ago
by claude+7.46%
fix issues
ffd090b
2 hours ago
by connortsui200%
remove q0 for most groups
5bd762b
2 hours ago
by connortsui20