A data engineer changes the liquid clustering keys on an existing table from `(created_date)` to `(created_date, region)` using `ALTER TABLE`. They then run a regular `OPTIMIZE`. Which concern is valid?

Question

Accepted Answer

C. Existing data files are not reclustered with the new keys by a regular `OPTIMIZE`; `OPTIMIZE FULL` must be run. When clustering keys are changed on a liquid-clustered table, subsequent regular `OPTIMIZE` runs will cluster only newly written data using the new keys. Existing data files that were already compacted under the old keys are not automatically reclustered. To force reclustering of all existing data with the updated keys, `OPTIMIZE FULL` (available in Databricks Runtime 16.0 and above) must be run explicitly.

A data engineer changes the liquid clustering keys on an existing table from `(created_date)` to `(created_date, region)` using `ALTER TABLE`. They then run a regular `OPTIMIZE`. Which concern is valid?

Related Questions

Discussion

A data engineer changes the liquid clustering keys on an existing table from (created_date) to (created_date, region) using ALTER TABLE. They then run a regular OPTIMIZE. Which concern is valid?

Related Questions

Discussion

A data engineer changes the liquid clustering keys on an existing table from `(created_date)` to `(created_date, region)` using `ALTER TABLE`. They then run a regular `OPTIMIZE`. Which concern is valid?