Hence, if a query is parallelized correctly it can benefit also from parallel I/O facilities and from the increased database cache that can reduce disk spoolings.