rdettai commented on issue #1363: URL: https://github.com/apache/arrow-datafusion/issues/1363#issuecomment-980070399
Interesting fact: Before #1010, the `read_partition` method was where 99% of the time was spent: ``` Query read_partition took 191.6 ms Query 6 iteration 0 took 193.1 ms Query read_partition took 178.3 ms Query 6 iteration 1 took 180.2 ms Query read_partition took 179.7 ms Query 6 iteration 2 took 181.2 ms ``` After #1010, this is not the case anymore, but `read_partition` itself is also slower: ``` Query read_partition took 253.3 ms Query 6 iteration 0 took 354.6 ms Query read_partition took 252.4 ms Query 6 iteration 1 took 354.6 ms Query read_partition took 258.9 ms Query 6 iteration 2 took 357.8 ms ``` So everything is slowed down, both on the thread where the reading occurs and in the Tokio threads... WeEEeeIiiiIrd 😅 -- 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]
