jtuglu1 commented on PR #18568:
URL: https://github.com/apache/druid/pull/18568#issuecomment-3355285601

   > * Scan query with limit is not cached. This is because the etag is 
populated after createResultLevelCachePopulator is called, resulting in a null 
resultLevelCachePopulator.   In case you want to take a look at that in this 
PR. I'm also fine if we address that later too. Btw this might be a common use 
case with the LIMIT / OFFSET to support pagination.
   
   I think this can be addressed in a follow-up PR. That would need to 
materialize in some form of "lazy" `ResultLevelCachePopulator` which would only 
be created once the etag was populated (e.g. after ClusteredClient is run).
   
   > * I think scan query intentionally did not implement caching. This is 
because the result set of scan query is typically large and will quickly fill 
the cache up causing higher cache miss rate. I think we should still make it 
possible to cache scan query but it should be disable by default given the 
above reasoning.
   
   I will add this to the default `druid...unCacheable=["scan"]` list. This way 
users need to opt-in to enable it.


-- 
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