drewdahlke opened a new issue #12108:
URL: https://github.com/apache/druid/issues/12108


   Hi, after extending druid with new aggregators I've found that 
AggregatorFactory.getMaxIntermediateSize(), used for memory estimation, has a 
huge impact on query performance. Smaller values are faster, but if your 
aggregator's intermediate results exceed the threshold then it fails. It's easy 
to dial that in with primitives, but some data sketches (eg frequent items) 
vary in size tremendously depending on the shape/quantity of input data it was 
fed. For folks extending druid to add new aggregators using complex types, are 
there any tricks to dealing with intermediate result sizes that greatly vary in 
size? Maybe this is a feature request to make that setting a soft 
limit/optimization rather than hard limit? 


-- 
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