[GitHub] Ben-Zvi opened a new pull request #1667: DRILL-7069: Moved version checks outside loops in transformBinaryInMetadataCache

[GitBox <gi...@apache.org>]

Ben-Zvi opened a new pull request #1667: DRILL-7069: Moved version checks outside loops in transformBinaryInMetadataCache
URL: https://github.com/apache/drill/pull/1667
 
 
       Performing the metadata version checks inside three level nested loops (in *transformBinaryInMetadataCache*) had a major performance impact (e.g., measured 1340 ms on a Parquet table of 4380 files, with 94 columns) for every relevant query.
      This code change extracted the metadata version checks from the loops. Running one loop set for V1, and another for the rest (the only difference for V3_3 is the 'needDecoding' flag).
      Also added debug code to report the runtime.
      Testing the modified code on the same Parquet table measured about 250 ms.
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services