alamb commented on issue #424:
URL: 
https://github.com/apache/arrow-datafusion/issues/424#issuecomment-848053786


   > if/when we make Ballista viable, we may want to dynamically optimize the 
query while it is executing,
   
   I do think dynamically changing the plan (or algorithms) based on actual 
execution experience is currently state of the art. I have often wondered if it 
is better done within the operators themselves (like maybe a join deciding to 
switch to sort-merge-join when it filled up its hash tables, or sampling both 
inputs to decide which was smaller. etc)
   
   I would have to think of the kinds of dynamic plan changes we might want to 
do 


-- 
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]


Reply via email to