yihua commented on issue #9026:
URL: https://github.com/apache/hudi/issues/9026#issuecomment-1627630057

   Based on my investigation, it looks like that the HFile does not contain 
duplicate entries.  If the metadata table is queried through Spark datasource 
directly after MDT compaction (i.e., no additional log file in the latest file 
slice), there is no duplicate.  The duplicate seems to be related to Spark 
datasource read on MDT only, not affecting the MDT read through 
`HoodieBackedTableMetadata` APIs.
   
   The performance is a separate issue we need to look into.


-- 
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