adriangb commented on PR #25338: URL: https://github.com/apache/datafusion/pull/25338#issuecomment-5733366325
Benchmark summary. Every run compares `e501c78` with the merge-base `64871d9`. [`projection_subquery`](https://github.com/apache/datafusion/pull/25338#issuecomment-5732942083), the suite this PR targets: | Query | `main` | this PR | | | --- | --- | --- | --- | | q01 bare `IN` | 246.7 ms | 1.13 ms | 218x | | q02 `COALESCE` over `IN` | 483.9 ms | 2.1 ms | 230x | | q03 correlated `IN`, equality | 4.2 ms | 1.62 ms | 2.6x | | q04 two `IN` columns | 410.5 ms | 1.90 ms | 216x | | q05 bare `NOT IN` | 251.6 ms | 1.11 ms | 228x | | q06 correlated `EXISTS` | 919 µs | 917 µs | 1.00x | | q07 correlated `IN`, non-equality | 99.5 ms | 100.8 ms | 1.01x | q06 and q07 are the controls. q07 keeps the three-join materialization, which this PR does not change, and https://github.com/apache/datafusion/issues/25336 covers it. The other suites show no change. Every query is "no change" in the min table: | Suite | Queries | Total `main` | Total this PR | | --- | --- | --- | --- | | [tpch sf1](https://github.com/apache/datafusion/pull/25338#issuecomment-5733284856) | 22 | 745.3 ms | 745.6 ms | | [tpcds sf1](https://github.com/apache/datafusion/pull/25338#issuecomment-5733300811) | 99 | 9293.3 ms | 9314.2 ms | | [clickbench partitioned](https://github.com/apache/datafusion/pull/25338#issuecomment-5733339285) | 43 | 20055.6 ms | 20059.5 ms | The mean tables move a few queries in each direction, which is inside the noise of this runner. -- 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]
