Exam Certified Data Engineer Professional topic 1 question 92 discussion

Shallow clone: only duplicates the metadata of the table being cloned; the data files of the table itself are not copied. These clones are cheaper to create but are not self-contained and depend on the source from which they were cloned as the source of data. If the files in the source that the clone depends on are removed, for example with VACUUM, a shallow clone may become unusable. Therefore, shallow clones are typically used for short-lived use cases such as testing and experimentation.

I was not sure whether B or D but somehow I think that running VACUUM comand does not invalidate SHALLOW CLONEs . I mean its just that the data referenced by the clone is no longer present. It can still happen that a SHALLOW CLONE is working even after a VACUUM command run on the cloned table (origin) . So B is not completely correct.

In Delta Lake, the VACUUM command deletes data files that are no longer referenced by a Delta table and are older than the retention threshold. When a table is cloned using SHALLOW CLONE, the clone references the same data files as the original table but creates a new transaction log. If VACUUM is run on the original table, it can delete data files that are still being referenced by the cloned table’s metadata, causing the cloned table to stop working. This is because the VACUUM command doesn’t know about the cloned table’s references to the data files. Therefore, it’s important to be cautious when running VACUUM on tables that have clones.

D is correct

Please refer: https://docs.databricks.com/en/delta/clone.html#what-are-the-semantics-of-delta-clone-operations

B is best