Reorder agg kernel dispatch, and have Combined use inner accumulators
#7889Comparing
ngates/aggregate-kernels
(842b03a
) withdevelop
(e160125
)+31%
- Improvement
- 24
- Untouched
- 1186
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
1210 total
take_search[(0.005, 1.0)]
vortex-array/benches/take_patches.rs+31%
3.1 ms2.3 ms
take_search[(0.005, 0.5)]
vortex-array/benches/take_patches.rs+31%
1.5 ms1.2 ms
take_search[(0.005, 0.1)]
vortex-array/benches/take_patches.rs+30%
320.3 µs247.2 µs
take_search[(0.01, 1.0)]
vortex-array/benches/take_patches.rs+29%
3.3 ms2.5 ms
take_search[(0.01, 0.5)]
vortex-array/benches/take_patches.rs+29%
1.6 ms1.3 ms
take_search[(0.005, 0.05)]
vortex-array/benches/take_patches.rs+28%
168.3 µs131.7 µs
take_search[(0.01, 0.1)]
vortex-array/benches/take_patches.rs+27%
341.2 µs268.1 µs
take_search[(0.01, 0.05)]
vortex-array/benches/take_patches.rs+26%
179.4 µs142.8 µs
take_search_chunked[(0.005, 1.0)]
vortex-array/benches/take_patches.rs+21%
3.5 ms2.9 ms
take_search_chunked[(0.005, 0.5)]
vortex-array/benches/take_patches.rs+21%
1.8 ms1.5 ms
take_search[(0.1, 1.0)]
vortex-array/benches/take_patches.rs+21%
4.3 ms3.5 ms
take_search[(0.1, 0.5)]
vortex-array/benches/take_patches.rs+20%
2.2 ms1.8 ms
take_search_chunked[(0.005, 0.1)]
vortex-array/benches/take_patches.rs+20%
369.1 µs307.2 µs
take_search_chunked[(0.01, 1.0)]
vortex-array/benches/take_patches.rs+20%
3.8 ms3.2 ms
take_search_chunked[(0.01, 0.5)]
vortex-array/benches/take_patches.rs+19%
1.9 ms1.6 ms
take_search_chunked[(0.005, 0.05)]
vortex-array/benches/take_patches.rs+19%
193.1 µs162.1 µs
take_search[(0.1, 0.1)]
vortex-array/benches/take_patches.rs+19%
458.6 µs385.6 µs
take_search_chunked[(0.01, 0.1)]
vortex-array/benches/take_patches.rs+19%
393.9 µs331.9 µs
take_search_chunked[(0.01, 0.05)]
vortex-array/benches/take_patches.rs+18%
206.2 µs175.2 µs
take_search[(0.1, 0.05)]
vortex-array/benches/take_patches.rs+17%
249.1 µs212.4 µs
take_search_chunked[(0.1, 1.0)]
vortex-array/benches/take_patches.rs+15%
4.7 ms4.1 ms
take_search_chunked[(0.1, 0.5)]
vortex-array/benches/take_patches.rs+15%
2.4 ms2.1 ms
take_search_chunked[(0.1, 0.1)]
vortex-array/benches/take_patches.rs+14%
504.2 µs442.1 µs
take_search_chunked[(0.1, 0.05)]
vortex-array/benches/take_patches.rs+13%
272 µs240.9 µs
true_count_arrow_buffer[128]
vortex-buffer/benches/vortex_bitbuffer.rs+8%
755.6 ns697.2 ns
Commits
Click on a commit to change the comparison rangeBase
develop
e160125
+31.26%
Reorder kernel dispatch, and have Combined use inner accumulators
7a65ca2
4 hours ago
by gatesn0%
Reorder kernel dispatch, and have Combined use inner accumulators
842b03a
4 hours ago
by gatesn