etseidl commented on PR #9945:
URL: https://github.com/apache/arrow-rs/pull/9945#issuecomment-4455067334
Running manually on an ARM instance
```
group branch
branch2 main2
----- ------
------- -----
decode metadata (wide) with schema 1.02 26.4±0.51ms
? ?/sec 1.00 26.0±0.43ms ? ?/sec 1.03 26.8±0.54ms
? ?/sec
decode metadata (wide) with skip PES 1.02 26.6±0.78ms
? ?/sec 1.09 28.3±0.39ms ? ?/sec 1.00 26.0±0.56ms
? ?/sec
decode metadata (wide) with skip all stats 1.12 32.4±0.26ms
? ?/sec 1.02 29.6±0.44ms ? ?/sec 1.00 28.9±0.46ms
? ?/sec
decode metadata (wide) with skip column stats 1.03 28.9±0.32ms
? ?/sec 1.00 28.1±0.49ms ? ?/sec 1.01 28.3±0.45ms
? ?/sec
decode metadata (wide) with skip size stats 1.07 31.7±0.43ms
? ?/sec 1.00 29.7±0.49ms ? ?/sec 1.04 30.7±0.68ms
? ?/sec
decode metadata (wide) with stats mask 1.02 25.8±0.56ms
? ?/sec 1.00 25.2±0.42ms ? ?/sec 1.03 26.0±0.50ms
? ?/sec
decode metadata with schema 1.02 4.1±0.01µs
? ?/sec 1.02 4.1±0.01µs ? ?/sec 1.00 4.0±0.02µs
? ?/sec
decode metadata with skip PES 1.00 6.9±0.04µs
? ?/sec 1.00 7.0±0.01µs ? ?/sec 1.00 6.9±0.04µs
? ?/sec
decode metadata with skip column stats 1.00 6.8±0.02µs
? ?/sec 1.01 6.9±0.09µs ? ?/sec 1.01 6.9±0.04µs
? ?/sec
decode metadata with stats mask 1.00 6.9±0.05µs
? ?/sec 1.00 6.9±0.02µs ? ?/sec 1.01 6.9±0.04µs
? ?/sec
decode parquet metadata 1.00 7.1±0.02µs
? ?/sec 1.00 7.2±0.02µs ? ?/sec 1.01 7.2±0.04µs
? ?/sec
decode parquet metadata (wide) 1.03 28.7±0.47ms
? ?/sec 1.00 27.8±0.35ms ? ?/sec 1.05 29.2±0.53ms
? ?/sec
decode parquet metadata no path_in_schema (wide) 1.04 28.4±0.54ms
? ?/sec 1.00 27.3±0.54ms ? ?/sec 1.03 28.2±0.49ms
? ?/sec
decode parquet metadata w/ size stats (wide) 1.04 35.6±0.42ms
? ?/sec 1.00 34.3±0.44ms ? ?/sec 1.04 35.6±0.49ms
? ?/sec
open(default) 1.00 7.8±0.05µs
? ?/sec 1.00 7.8±0.02µs ? ?/sec 1.00 7.8±0.04µs
? ?/sec
open(page index) 1.00 132.0±0.22µs
? ?/sec 1.00 132.0±0.18µs ? ?/sec 1.00 132.0±0.23µs
? ?/sec
```
picked the best run of the current main branch vs 2 runs of this branch. I
think we can conclude it's just noise, and if there is a difference it's not
really showing up strongly (although on x86 there's a definite advantage to
this branch).
Thanks @alamb!
--
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]
