Comparing
ad/cuda-copy-benchmarks
(fc4cda4
) withdevelop
(0b790d2
)0%
- Untouched
- 1206
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
1206 total
cuda/dict_256vals_bp8bw_codes/dispatch_u32[100M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::cuda+5%
1.1 ms1.1 ms
true_count_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+5%
648.9 ns619.7 ns
true_count_arrow_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
965.6 ns936.4 ns
iter_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
981.1 ns951.9 ns
true_count_arrow_buffer[2048]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
1 µs1 µs
set_indices_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
1 µs1 µs
cuda/alp_for_bp_6bw_f32/dispatch_f32[100M]
vortex-cuda/benches/dynamic_dispatch_cuda.rs::benches::benchmark_dynamic_dispatch::cuda+2%
1.1 ms1.1 ms
decompress[u64, (1000, 16)]
encodings/runend/benches/run_end_compress.rs+2%
22 µs21.6 µs
decompress[u8, (1000, 16)]
encodings/runend/benches/run_end_compress.rs+1%
17.5 µs17.2 µs
true_count_arrow_buffer[16384]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
2 µs2 µs
decompress[u32, (10000, 1024)]
encodings/runend/benches/run_end_compress.rs+1%
34.2 µs33.8 µs
set_indices_vortex_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
2.3 µs2.2 µs
bitwise_and_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
4.6 µs4.6 µs
bitwise_or_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
4.7 µs4.6 µs
take_indices[(1000, 256)]
encodings/runend/benches/run_end_compress.rs+1%
34.6 µs34.2 µs
cuda/bitpacked_patched_u16/unpack/10%[100M]
vortex-cuda/benches/bitpacked_cuda.rs::benches::benchmark_bitunpack_with_patches::cuda+1%
721.8 µs714 µs
decompress[u64, (1000, 4)]
encodings/runend/benches/run_end_compress.rs+1%
28.2 µs27.9 µs
decompress[u32, (10000, 256)]
encodings/runend/benches/run_end_compress.rs+1%
36.6 µs36.3 µs
encode_varbinview[(10000, 32)]
vortex-array/benches/dict_compress.rs+1%
1 ms1 ms
decompress_alp[f32, (1000, 0.01, 1.0)]
encodings/alp/benches/alp_compress.rs+1%
12.7 µs12.6 µs
take_indices[(1000, 16)]
encodings/runend/benches/run_end_compress.rs+1%
38.3 µs38 µs
decompress_alp[f32, (1000, 0.1, 1.0)]
encodings/alp/benches/alp_compress.rs+1%
13.7 µs13.6 µs
decompress_fsst[(1000, 16, 4)]
encodings/fsst/benches/fsst_compress.rs+1%
82.6 µs82 µs
set_indices_vortex_buffer[2048]
vortex-buffer/benches/vortex_bitbuffer.rs+1%
3.6 µs3.6 µs
decompress[u16, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+1%
17 µs16.9 µs
Commits
Click on a commit to change the comparison rangeBase
develop
0b790d2
-0.02%
bench: CUDA host-to-device copy modes
fc4cda4
2 days ago
by 0ax1