clintropolis commented on PR #13710:
URL: https://github.com/apache/druid/pull/13710#issuecomment-1404750983

   it would probably be worth checking that query performance for realtime 
tasks is not negatively impacted too much by this change (especially from 
multiple concurrent threads). There is only one writer thread but potentially 
many reader threads, as new dimension selectors are created on top of this 
dictionary during realtime queries.
   
   For batch ingestion tasks, it might be worth using a separate 
`DimensionDictionary` implementation that has no locks. I use a similar thing 
for the `NestedDataColumnIndexer`, `ComparatorDimensionDictionary`, which is 
basically a copy of `DimensionDictionary` except uses a `Comparator` instead of 
requiring a `Comparable`, and doesn't have any locks since the nested column 
indexer doesn't currently use the dictionary it is building for realtime 
queries.


-- 
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