CACHE SELECT (Databricks SQL)
Caches the data accessed by the specified simple
SELECT query in the Delta cache. You can choose a subset of columns to be cached by providing a list of column names and choose a subset of rows by providing a predicate. This enables subsequent queries to avoid scanning the original files as much as possible. This construct is applicable only to Parquet tables. Views are also supported, but the expanded queries are restricted to the simple queries, as described above.
CACHE SELECT column_name [, ...] FROM table_name [ WHERE boolean_expression ]
See Delta and Apache Spark caching for the differences between the Delta cache and the Apache Spark cache.
CACHE SELECT * FROM boxes CACHE SELECT width, length FROM boxes WHERE height=3