When integrating object storage with Autonomous Database, how effectively does Oracle use the files in...

When Autonomous Database integrates with object storage (e.g., Oracle Cloud Infrastructure Object Storage), it optimizes data access. The correct answer is:

Scan partition tables (B):Oracle leverages partition pruning when scanning tables stored in object storage. This means it only scans the relevant partitions based on query predicates, significantly reducing I/O and improving performance. This is particularly effective for partitioned tables stored in formats like Parquet or ORC, which support partitioning.

The incorrect options are:

Prune columns in parquet files (A):While column pruning is possible with columnar formats like Parquet, the primary optimization highlighted in Oracle’s documentation for object storage integration is partition pruning, not column pruning alone.

Scan hybrid partition tables (C):“Hybrid partition tables” is not a standard term in this context. Oracle supports external partitioned tables, but the focus is on scanning partitioned tables, not a hybrid-specific optimization.

Prune columns in CSV files (D):CSV files are row-based and do not natively support column pruning, making this less effective compared to partition scanning.

Partition pruning is a key optimization for external tables in object storage, enhancing query efficiency.