jp0317 commented on PR #39818:
URL: https://github.com/apache/arrow/pull/39818#issuecomment-1943734729

   changed to batch-skip instead of resizing buffer, as per @pitrou suggested 
(I had to reduce the buffer size to 64 Bytes to make the tests actually doing 
multiple batches of skips, because the number of buffered rows is always less 
that 
[kMinLevelBatchSize](https://github.com/apache/arrow/blob/main/cpp/src/parquet/column_reader.cc#L1369)).
 
   
   for BatchSize less than 512, the improvement is still there. For BatchSize 
`1000` and `5000`, the median time slightly increased to `78554904` and  
`74765553`.


-- 
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]

Reply via email to