nealrichardson commented on pull request #11373:
URL: https://github.com/apache/arrow/pull/11373#issuecomment-942723128


   > The only query I would expect to see a large speedup on is a head with no 
arrange
   
   head with no arrange doesn't use top-k, there's nothing to "top". So it's 
using the Scanner::Head method on top of the RecordBatchReader we get back from 
the ExecPlan
   
   I would have expected top-k would at least work better in memory constrained 
cases because it doesn't have to keep around more than k rows of the result 
set, but maybe I don't understand how it's implemented.


-- 
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: github-unsubscr...@arrow.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Reply via email to