Re: [I] Hash partitioning not falling back to Spark for high precision decimal partition keys [datafusion-comet]

["andygrove (via GitHub)" <gi...@apache.org>]

andygrove commented on issue #3079:
URL: https://github.com/apache/datafusion-comet/issues/3079#issuecomment-6196055986

   #6005 fixed this. Native shuffle now declines a hash key with a decimal wider than 18 digits, nested fields included, whenever it hashes into more than one partition, so those shuffles use Spark's partitioner and stay copartitioned with inputs that Spark partitioned. That covers the mixed native and columnar shuffle join above, the cached relation case, and bucketed tables, which #6005 added a test for. #5994 tracks making the native hash match Spark's so these keys can use native shuffle again.


-- 
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: github-unsubscribe@datafusion.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscribe@datafusion.apache.org
For additional commands, e-mail: github-help@datafusion.apache.org