gautamworah96 commented on issue #12576:
URL: https://github.com/apache/lucene/issues/12576#issuecomment-1809159446

   I am not actively working on this problem as of now. 
   I am still in the process of figuring out what would be the correct thing to 
test/do here as a first step. Jotting down some thoughts that I had.
   
   1. We could try the `Collect into a sparse structure at first, and if it 
gets too big, switch to dense.` experiment. It adds a costs to the `setValue` 
call when it will have to suddenly switch to a dense structure. But that should 
be fine? It is all within the same query. So no cases of `a single query in a 
sea of queries suddenly being too slow`. What heuristic on when to switch to 
use here?
   2. Do we need to make the `setValue` call operation thread safe? How will 
that work with the dynamic changing?
   
   `luceneutil` needs to be hacked to record JFR stats for determining if this 
change improves the heap allocation or not.


-- 
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