DeltaLake vs parquet file

[dridk]

Hi,

I don't see any real advantage in using delta lake instead of simple parquet files. but I may be wrong.

My ETL process consists of batch extracting of data partitioned by month. So 1 parquet file per month on a local storage.

-If I have a problem with a corrupt file or need to update data, I delete the parquet file and regenerate it.
-Performances are greate with Partition, stats and duckdb.

So I don't need an ACID property or history.

Is there something I've missed?

[ion-elgreco]

@dridk delta lake allows you to z-order your data so you can do faster filter operations on your table.