theirix commented on PR #19980:
URL: https://github.com/apache/datafusion/pull/19980#issuecomment-3801568160

   > I think we'd have to see benchmarks from the microbenchmark of `left` 
itself; the benchmark queries from the bot don't use this function I think? 
(Unless its used internally in joins, etc.)
   
   Agree, here are the results of a more targeted `left` benchmark (they don't 
cover string views):
   
   ```
   left size=1024/positive n/1024
                           time:   [28.333 µs 29.190 µs 30.124 µs]
                           change: [−75.427% −74.733% −74.087%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 2 outliers among 100 measurements (2.00%)
     2 (2.00%) high mild
   left size=1024/negative n/1024
                           time:   [24.780 µs 25.203 µs 25.711 µs]
                           change: [−88.075% −87.696% −87.324%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 1 outliers among 100 measurements (1.00%)
     1 (1.00%) high mild
   
   left size=4096/positive n/4096
                           time:   [97.199 µs 97.773 µs 98.422 µs]
                           change: [−82.227% −80.928% −79.732%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 8 outliers among 100 measurements (8.00%)
     6 (6.00%) high mild
     2 (2.00%) high severe
   left size=4096/negative n/4096
                           time:   [168.66 µs 173.51 µs 178.90 µs]
                           change: [−80.799% −80.005% −79.258%] (p = 0.00 < 
0.05)
                           Performance has improved.
   Found 2 outliers among 100 measurements (2.00%)
     1 (1.00%) high mild
     1 (1.00%) high severe
   ```


-- 
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]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to