Hi,

You mentioned that you have tried Z-ordering but it didn't impact the performance. Z-ordering is a technique that co-locates related information in the same set of files. It works best when the data is filtered by the column specified in the Z-ordering. If your queries are not filtering on the Z-ordered column, you might not see a significant performance improvement.
https://docs.databricks.com/en/delta/data-skipping.html

https://www.databricks.com/blog/2020/05/29/adaptive-query-execution-speeding-up-spark-sql-at-runtime...

https://www.databricks.com/blog/2020/04/30/faster-sql-queries-on-delta-lake-with-dynamic-file-prunin...