ariel-miculas opened a new issue, #23559:
URL: https://github.com/apache/datafusion/issues/23559

   ### Describe the bug
   
   Basically in multi level merge: 
https://github.com/apache/datafusion/blob/b7e7b514e9f93c0a013286511a36e86ffcf2d662/datafusion/physical-plan/src/sorts/multi_level_merge.rs#L456-L465
   
   the reservation is multiplied by buffer_len.
   
   This buffer_size is later on used to read the spill files as a stream:
   
https://github.com/apache/datafusion/blob/b7e7b514e9f93c0a013286511a36e86ffcf2d662/datafusion/physical-plan/src/sorts/multi_level_merge.rs#L329-L339
   
   read_spill_as_stream calls spawn_buffered with 
self.batch_read_buffer_capacity:
   
https://github.com/apache/datafusion/blob/b7e7b514e9f93c0a013286511a36e86ffcf2d662/datafusion/physical-plan/src/spill/spill_manager.rs#L194
   
   The maximum amount of RecordBatches produced by this spawn_buffered call:
   
https://github.com/apache/datafusion/blob/739e1471b262e9e3b2e26089d4678ad4c054c399/datafusion/physical-plan/src/common.rs#L191-L208
   
   is actually `buffer + 1`.
   Example with buffer = 1:
   RecordBatch is produced, channel capacity is 1, sender.send succeeds, 1 
RecordBatch in memory
   RecordBatch is produced, channel is full, sender.send blocks, but the 
RecordBatch has already been produced => 2 RecordBatches in memory
   
   This is an issue especially with large Record batches
   
   ### To Reproduce
   
   _No response_
   
   ### Expected behavior
   
   Since buffer needs to be at least one, get a permit before producing the 
RecordBatch: https://docs.rs/tokio/latest/tokio/sync/mpsc/struct.Permit.html
   Update all the callers to use buffer + 1 in order to preserve behavior
   
   ### Additional context
   
   _No response_


-- 
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