HippoBaro commented on PR #9831:
URL: https://github.com/apache/arrow-rs/pull/9831#issuecomment-4376865540
Hum, I reran the above benchmarks locally on my laptop, both before and
after rebasing onto `main`, and I can’t reproduce the slowdown. I get very
consistent overlapping results (`list_primitive_non_null/parquet_2`):
main 0e478d8f5: [279.28 ms 280.83 ms 282.68 ms]
branch ccb0ef114: [277.20 ms 278.52 ms 280.15 ms]
I’m not sure I trust this particular benchmark runner result. For example,
the worst offender in your list, `list_primitive_non_null/parquet_2`, was flat
in [the latest benchmark
run](https://github.com/apache/arrow-rs/pull/9831#issuecomment-4366323483):
list_primitive_non_null/parquet_2 1.01 311.3±6.23ms 1748.4 MB/sec
1.00 308.4±13.64ms 1764.9 MB/sec
So I strongly suspect this is benchmark noise 🤷
--
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]
