siddharthteotia opened a new pull request #5176: Make text index query cache a 
configurable option
URL: https://github.com/apache/incubator-pinot/pull/5176
 
 
   During design of text index feature, we had done some heap overhead 
experiments. As part of that we had determined that while Lucene's internal 
query cache helps with performance significantly for repeatable queries, it 
adds heap overhead. Therefore, it was disabled during index loading.
   
   It is worthwhile to make this configurable on per-index basis for users to 
enable it. Of course, they need to know the downside of more heap overhead 
(which could potentially negate the perf improvements due to more GC).
   
   As part of ongoing internal user acceptance testing, we learned that most of 
the queries are repeatable. The UI will more or less keep the text search 
filter constant and tweak the other filters. For such cases, it is good to see 
the performance improvements by enabling the query cache and if the heap 
overhead is not significantly high, user might want to keep it enabled.
   
   Note: By default it is still disabled. So nothing really changed in the 
existing behavior.

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to