cuda FFOR + perfetto logging
#6588Comparing
aduffy/fix-gpu-scan
(d357c4f
) withdevelop
(55c7508
)+1%
- Untouched
- 977
- Skipped
- 1426
Benchmarks
2403 total
decompress[u8, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+5%
17.2 µs16.4 µs
set_indices_vortex_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+5%
1.3 µs1.2 µs
decompress[u16, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+4%
19.3 µs18.5 µs
decompress[u32, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+4%
20 µs19.2 µs
decompress[u8, (1000, 16)]
encodings/runend/benches/run_end_compress.rs+4%
19.8 µs19 µs
decompress[u32, (1000, 16)]
encodings/runend/benches/run_end_compress.rs+4%
23.6 µs22.7 µs
decompress[u16, (1000, 16)]
encodings/runend/benches/run_end_compress.rs+4%
22.4 µs21.6 µs
null_count_run_end[(10000, 1024, 0.5)]
encodings/runend/benches/run_end_null_count.rs+4%
41.3 µs39.8 µs
null_count_run_end[(10000, 1024, 0.1)]
encodings/runend/benches/run_end_null_count.rs+4%
41.4 µs39.9 µs
map_each[BufferMut<i32>, 128]
vortex-buffer/benches/vortex_buffer.rs+4%
828.9 ns799.7 ns
null_count_run_end[(10000, 256, 0.5)]
encodings/runend/benches/run_end_null_count.rs+3%
43.6 µs42.2 µs
null_count_run_end[(10000, 256, 0.1)]
encodings/runend/benches/run_end_null_count.rs+3%
43.8 µs42.3 µs
decompress[u8, (10000, 1024)]
encodings/runend/benches/run_end_compress.rs+3%
23.3 µs22.5 µs
chunked_dict_primitive_canonical_into[f64, (1000, 10, 10)]
vortex-array/benches/chunked_dict_builder.rs+3%
136.8 µs132.4 µs
decompress[u8, (10000, 256)]
encodings/runend/benches/run_end_compress.rs+3%
24.6 µs23.8 µs
decompress[u8, (1000, 4)]
encodings/runend/benches/run_end_compress.rs+3%
25.4 µs24.6 µs
decompress[u64, (1000, 256)]
encodings/runend/benches/run_end_compress.rs+3%
21.6 µs20.9 µs
decompress[u16, (10000, 1024)]
encodings/runend/benches/run_end_compress.rs+3%
27.9 µs27 µs
decode_varbinview[(1000, 4)]
vortex-array/benches/dict_compress.rs+3%
33.9 µs32.9 µs
decode_varbin[(1000, 8)]
vortex-array/benches/dict_compress.rs+3%
34 µs33.1 µs
decode_varbinview[(1000, 2)]
vortex-array/benches/dict_compress.rs+3%
34.4 µs33.4 µs
decode_varbin[(1000, 2)]
vortex-array/benches/dict_compress.rs+3%
34.3 µs33.4 µs
decode_varbin[(1000, 32)]
vortex-array/benches/dict_compress.rs+3%
34.8 µs33.8 µs
decode_varbin[(1000, 4)]
vortex-array/benches/dict_compress.rs+3%
34.7 µs33.7 µs
decode_varbinview[(1000, 8)]
vortex-array/benches/dict_compress.rs+3%
34.6 µs33.6 µs
Commits
Click on a commit to change the comparison rangeBase
develop
55c7508
+0.51%
use tracing-perfetto instead
3ed9fdf
12 hours ago
by a10y0%
clang-format
d0119f2
12 hours ago
by a10y0%
fix clang-format
d357c4f
11 hours ago
by a10y