WIP: gpu backed scan
#6192Comparing
ji/os/gpu-scan
(1674b3e
) withaduffy/filter-decimals
(cc6afa5
)-25%
- Improvement
- 1
- Regression
- 5
- Untouched
- 1173
- Skipped
- 1323
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
2502 total
u16_FoR[10M]
vortex-cuda/benches/for_cuda.rs::benches::benchmark_for_cuda::FoR_cuda_u16-25%
8.1 µs10.9 µs
1M_10pct[100000]
vortex-cuda/benches/filter_cuda.rs::benches::benchmark_filter_cuda::Filter_cuda_i32-24%
22.8 µs30 µs
1M_50pct[500000]
vortex-cuda/benches/filter_cuda.rs::benches::benchmark_filter_cuda::Filter_cuda_i32-12%
24 µs27.2 µs
1M_90pct[1000000]
vortex-cuda/benches/filter_cuda.rs::benches::benchmark_filter_cuda::Filter_cuda_i32-11%
30.2 µs34 µs
1M_10pct[100000]
vortex-cuda/benches/filter_cuda.rs::benches::benchmark_filter_cuda::Filter_cuda_f64-11%
44.2 µs49.4 µs
u64_values_u8_codes[10M]
vortex-cuda/benches/dict_cuda.rs::benches::benchmark_dict_cuda::Dict_cuda_u64_u8+17%
259.4 µs222.5 µs
new_bp_prim_test_between[i32, 2048]
encodings/fastlanes/benches/compute_between.rs::bitpack+5%
38 µs36.3 µs
chunked_dict_primitive_canonical_into[u32, (1000, 100, 10)]
vortex-array/benches/chunked_dict_builder.rs+3%
95.5 µs92.9 µs
chunked_dict_primitive_canonical_into[u32, (1000, 10, 10)]
vortex-array/benches/chunked_dict_builder.rs+3%
94.1 µs91.7 µs
decode_varbinview[(1000, 32)]
vortex-array/benches/dict_compress.rs+3%
28.5 µs27.8 µs
set_indices_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.2 µs1.2 µs
value_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.3 µs1.3 µs
decode_primitives[i64, (1000, 2)]
vortex-array/benches/dict_compress.rs+2%
22.4 µs21.9 µs
dict_canonicalize_uniform[16, 1000]
vortex-array/benches/take_primitive.rs+2%
20.2 µs19.8 µs
10M_90pct[10000000]
vortex-cuda/benches/filter_cuda.rs::benches::benchmark_filter_cuda::Filter_cuda_i32+2%
199.7 µs195.5 µs
dict_canonicalize_uniform[256, 1000]
vortex-array/benches/take_primitive.rs+2%
20.9 µs20.5 µs
append_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.5 µs1.4 µs
set_indices_arrow_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
2.9 µs2.9 µs
encode_primitives[f32, (1000, 4)]
vortex-array/benches/dict_compress.rs+2%
121.2 µs118.8 µs
dict_canonicalize_zipfian[256, 1000]
vortex-array/benches/take_primitive.rs+2%
20.8 µs20.4 µs
dict_canonicalize_zipfian[2048, 1000]
vortex-array/benches/take_primitive.rs+2%
23.6 µs23.2 µs
chunked_varbinview_opt_canonical_into[(10, 1000)]
vortex-array/benches/chunk_array_builder.rs+2%
6.8 ms6.7 ms
varbinview_zip_fragmented_mask
vortex-array/benches/varbinview_zip.rs+2%
33.9 ms33.3 ms
filter_selectivity[u64, 0.01]
vortex-compute/benches/filter_buffer_mut.rs+2%
3.4 µs3.3 µs
chunked_dict_primitive_canonical_into[u32, (1000, 10, 100)]
vortex-array/benches/chunked_dict_builder.rs+2%
753.5 µs740.6 µs
Commits
Click on a commit to change the comparison rangeBase
aduffy/filter-decimals
cc6afa5
-25.14%
test with struct/varbin passing
1674b3e
7 hours ago
by a10y