CUDA execution for external patches
#7376Comparing
aduffy/patched-fuse
(8796e8c
) withdevelop
(256a029
)0%
- Untouched
- 1122
- Skipped
- 1455
Benchmarks
2577 total
take_map[(0.1, 1.0)]
vortex-array/benches/take_patches.rs+9%
1.8 ms1.7 ms
take_map[(0.1, 0.5)]
vortex-array/benches/take_patches.rs+8%
1,052.6 µs974.7 µs
take_map[(0.1, 0.1)]
vortex-array/benches/take_patches.rs+5%
468.1 µs447.3 µs
binary_search_vortex
vortex-array/benches/search_sorted.rs+4%
727.2 ns698.1 ns
encode_varbin[(10000, 8)]
vortex-array/benches/dict_compress.rs+4%
1,027.7 µs987.8 µs
true_count_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+4%
755.6 ns726.4 ns
iter_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+4%
861.1 ns831.9 ns
take_map[(0.1, 0.05)]
vortex-array/benches/take_patches.rs+3%
394 µs381.2 µs
true_count_vortex_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
927.5 ns898.3 ns
value_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
1,011.1 ns981.9 ns
true_count_vortex_buffer[2048]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
1,027.5 ns998.3 ns
set_indices_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+3%
1.1 µs1 µs
set_indices_arrow_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
2.4 µs2.4 µs
push_vortex_buffer[128]
vortex-buffer/benches/vortex_buffer.rs+2%
1.2 µs1.2 µs
slice_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.3 µs1.3 µs
slice_vortex_buffer[65536]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.3 µs1.3 µs
slice_vortex_buffer[1024]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.3 µs1.3 µs
slice_vortex_buffer[2048]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.3 µs1.3 µs
slice_vortex_buffer[16384]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
1.3 µs1.3 µs
append_n_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+2%
2.8 µs2.8 µs
decompress_fsst[(1000, 4, 4)]
encodings/fsst/benches/fsst_compress.rs+2%
65.4 µs64.1 µs
take_struct_sequential_indices
vortex-array/benches/take_struct.rs+2%
49.8 µs48.9 µs
decompress_fsst[(1000, 4, 8)]
encodings/fsst/benches/fsst_compress.rs+2%
67.2 µs66 µs
push_n_vortex_buffer[u32, 128]
vortex-buffer/benches/vortex_buffer.rs+2%
1.8 µs1.7 µs
dict_canonicalize_uniform[16, 1000]
vortex-array/benches/take_primitive.rs+2%
19.8 µs19.5 µs
Commits
Click on a commit to change the comparison rangeBase
develop
256a029
-0.03%
CUDA execute Patched
721fa3d
2 days ago
by a10y+0.02%
add unit test
681c73e
2 days ago
by a10y0%
copy to device before assembling device patches
8796e8c
2 days ago
by a10y