adriangb commented on issue #20324:
URL: https://github.com/apache/datafusion/issues/20324#issuecomment-3893747311

   > Hey [@alamb](https://github.com/alamb) I was looking into Q3 and found 
that the aggregate dynamic filters were being pushed down to RowFilter. We also 
start with zero selectivity which in turn qualifies more files for scanning 
than with filter pushdown turned off.
   > 
   > some explain analyze stats:
   > 
   >     * `pushdown_rows_pruned=23.14 M` (none without filter)
   > 
   >     * `files_ranges_pruned_statistics=113 total → 31 matched` (6 matched 
without filter)
   > 
   > 
   > perhaps for aggregate queries we could avoid evaluating rowgroups! Let me 
know what you think!
   
   Q3 is `SELECT AVG("UserID") FROM hits;`
   
   As far as I know no dynamic filters should be generated (at least after your 
fix in https://github.com/apache/datafusion/pull/20279). Q3 is also not one of 
the queries that gets slower.


-- 
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