askoa opened a new pull request, #3679:
URL: https://github.com/apache/arrow-rs/pull/3679
# Which issue does this PR close?
# Rationale for this change
I recently noticed an issue with the parameters in the benchmark
`primitive_run_take`. The benchmark was not taking indices across the entire
logical array but rather constrained to the length of take array. I updated
the parameters to take indices across the entire logical array. There is
performance regression because more values are taken from the array. I have
furnished the performance differences below
<details>
<summary><h4>Benchmark results</h4></summary>
```
primitive_run_take/(run_array_len:512, physical_array_len:64, take_len:512)
time: [23.347 µs 23.508 µs 23.713 µs]
change: [-1.5238% +0.0502% +1.6099%] (p = 0.95 >
0.05)
No change in performance detected.
Found 17 outliers among 100 measurements (17.00%)
3 (3.00%) high mild
14 (14.00%) high severe
primitive_run_take/(run_array_len:512, physical_array_len:128, take_len:512)
time: [23.477 µs 23.616 µs 23.804 µs]
change: [-1.2725% +0.2827% +1.9661%] (p = 0.75 >
0.05)
No change in performance detected.
Found 11 outliers among 100 measurements (11.00%)
1 (1.00%) high mild
10 (10.00%) high severe
primitive_run_take/(run_array_len:1024, physical_array_len:256, take_len:512)
time: [23.653 µs 23.778 µs 23.975 µs]
change: [-1.3315% +0.3299% +2.0635%] (p = 0.72 >
0.05)
No change in performance detected.
Found 13 outliers among 100 measurements (13.00%)
3 (3.00%) high mild
10 (10.00%) high severe
primitive_run_take/(run_array_len:1024, physical_array_len:256,
take_len:1024)
time: [47.861 µs 48.179 µs 48.621 µs]
change: [-0.3026% +1.4710% +3.2931%] (p = 0.10 >
0.05)
No change in performance detected.
Found 8 outliers among 100 measurements (8.00%)
4 (4.00%) high mild
4 (4.00%) high severe
primitive_run_take/(run_array_len:2048, physical_array_len:512, take_len:512)
time: [24.226 µs 24.287 µs 24.356 µs]
change: [+0.7574% +3.6336% +7.6136%] (p = 0.02 <
0.05)
Change within noise threshold.
Found 11 outliers among 100 measurements (11.00%)
3 (3.00%) high mild
8 (8.00%) high severe
primitive_run_take/(run_array_len:2048, physical_array_len:512,
take_len:1024)
time: [53.774 µs 54.093 µs 54.615 µs]
change: [+12.948% +15.613% +18.678%] (p = 0.00 <
0.05)
Performance has regressed.
Found 13 outliers among 100 measurements (13.00%)
2 (2.00%) high mild
11 (11.00%) high severe
primitive_run_take/(run_array_len:4096, physical_array_len:1024,
take_len:512)
time: [25.297 µs 25.365 µs 25.440 µs]
change: [+4.7053% +7.0111% +9.3052%] (p = 0.00 <
0.05)
Performance has regressed.
Found 13 outliers among 100 measurements (13.00%)
5 (5.00%) high mild
8 (8.00%) high severe
primitive_run_take/(run_array_len:4096, physical_array_len:1024,
take_len:1024)
time: [59.299 µs 59.752 µs 60.351 µs]
change: [+21.297% +23.389% +25.448%] (p = 0.00 <
0.05)
Performance has regressed.
Found 16 outliers among 100 measurements (16.00%)
8 (8.00%) high mild
8 (8.00%) high severe
```
</details>
# What changes are included in this PR?
# Are there any user-facing changes?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
