aho135 commented on issue #20326:
URL: https://github.com/apache/druid/issues/20326#issuecomment-5655576164
@FrankChen021 The query itself has a simple pattern like:
```
WITH datapoints AS (
SELECT * FROM <datasource>
WHERE dim1 IN (v1, v2, ... vn)
)
```
I ran the jmh benchmarks with some small query modifications to include the
CTE but it did not have any material impact on planning time.
I did some more analysis around the degradation window and actually found
that even some simple queries had extremely long planning times. Turns out the
long planning times for those were actually just a side effect of stop the
world GC pauses caused by a flurry of queries with large IN clauses. The
planning timeout should help alleviate some of the GC pressure in this
scenario, but curious if you have any other thoughts on this scenario
--
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]
