corleyma commented on issue #2159:
URL: 
https://github.com/apache/iceberg-python/issues/2159#issuecomment-3032961475

   I think @Anton-Tarazi's original point -- creating a bunch of (Python 
object) filter expressions for every row in a large dataframe is going to be 
slow, and we do that before any real i/o happens -- this is done so that, as 
Fokko says, 
   >  we filter on only relevant files based on the Iceberg metadata.
   
   My guess is for tables of a certain size/layout you'd do better to just skip 
this metadata-based file pruning given the expensive python object creations.
   
   Honestly, I think it would be a better use of community resources to invest 
more in the iceberg-rust/datafusion path so that the bulk of this logic can be 
moved out of pure python, than to spend a ton of effort trying to optimize the 
Python implementation.


-- 
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