Comparing
ct/fix-cosine-denorm-opt
(3a54a19
) withdevelop
(0bb712b
)-42%
- Improvement
- 1
- Regression
- 4
- Untouched
- 1184
- Skipped
- 9
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
1198 total
for[10M_u8]
vortex-cuda/benches/for_cuda.rs::benches::benchmark_for::for_cuda-42%
73.8 µs127.4 µs
for[10M_u16]
vortex-cuda/benches/for_cuda.rs::benches::benchmark_for::for_cuda-39%
95.5 µs157.3 µs
new_bp_prim_test_between[i64, 32768]
encodings/fastlanes/benches/compute_between.rs::bitpack-25%
176.4 µs235.3 µs
mix[0%_in/100%_out]
vortex-cuda/benches/throughput_cuda.rs::benches::benchmark_transfer_throughput::transfer_throughput_cuda-18%
227.3 µs278.5 µs
dynamic_dispatch_u32[10M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::dict_nowiden_u32_to_u32+54%
162.8 µs105.5 µs
new_alp_prim_test_between[f32, 32768]
encodings/fastlanes/benches/compute_between.rs::alp+9%
182.9 µs168.2 µs
new_alp_prim_test_between[f32, 2048]
encodings/fastlanes/benches/compute_between.rs::alp+3%
61.9 µs60.1 µs
dynamic_dispatch_f32[10M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::alp_for_bp_6bw_f32+3%
115.7 µs112.6 µs
dynamic_dispatch_u32[10M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::runend_100runs+2%
95.2 µs93.2 µs
alp_decode[0%]
vortex-cuda/benches/alp_cuda.rs::benches::benchmark_alp_decode::alp_cuda_f64+2%
3.4 ms3.3 ms
alp_decode[10%]
vortex-cuda/benches/alp_cuda.rs::benches::benchmark_alp_decode::alp_cuda_f32+1%
4.5 ms4.4 ms
new_bp_prim_test_between[i16, 2048]
encodings/fastlanes/benches/compute_between.rs::bitpack+1%
51.8 µs51.1 µs
new_bp_prim_test_between[i32, 2048]
encodings/fastlanes/benches/compute_between.rs::bitpack+1%
53.6 µs52.9 µs
mix[50%_in/50%_out]
vortex-cuda/benches/throughput_cuda.rs::benches::benchmark_transfer_throughput::transfer_throughput_cuda+1%
397.3 µs392.2 µs
for[10M_u8]
vortex-cuda/benches/for_cuda.rs::benches::benchmark_ffor::ffor_cuda+1%
74.6 µs73.7 µs
dynamic_dispatch_u32[10M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::for_bitpacked_6bw+1%
115.7 µs114.7 µs
alp_decode[0%]
vortex-cuda/benches/alp_cuda.rs::benches::benchmark_alp_decode::alp_cuda_f32+1%
2.5 ms2.5 ms
dynamic_dispatch_u32[10M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::dict_256vals_bp8bw_codes+1%
126 µs124.9 µs
decompress_fsst[(1000, 4, 8)]
encodings/fsst/benches/fsst_compress.rs+1%
64.4 µs64.1 µs
take_map[(0.1, 0.05)]
vortex-array/benches/take_patches.rs0%
389.7 µs387.9 µs
decompress_utf8[(1000, 4)]
encodings/runend/benches/run_end_compress.rs0%
37.3 µs37.2 µs
bitunpack_patched[1%]
vortex-cuda/benches/bitpacked_cuda.rs::benches::benchmark_bitunpack_with_patches::bitunpack_cuda_patched_u80%
383.4 µs381.8 µs
alp_decode[1%]
vortex-cuda/benches/alp_cuda.rs::benches::benchmark_alp_decode::alp_cuda_f320%
4.8 ms4.8 ms
pushdown_compare[(1000, 64, 4)]
encodings/fsst/benches/fsst_compress.rs0%
341.4 µs340.1 µs
bitunpack_patched[10%]
vortex-cuda/benches/bitpacked_cuda.rs::benches::benchmark_bitunpack_with_patches::bitunpack_cuda_patched_u80%
483 µs481.3 µs
Commits
Click on a commit to change the comparison rangeBase
develop
0bb712b
-42.11%
fix cosine similarity optimization bug
3a54a19
5 hours ago
by connortsui20