Which two statements are true about the use of direct path loads when selecting from external tables in a database on a Database Machine?
A. INSERT INTO . . . SELECT FROM statements, executed serially, which select from external tables,
require the APPEND hint to use direct path loading.
B. CREATE TABLE . . . AS SELECT statements, which select from external tables, attempt to use in
direct path loading automatically.
C. CREATE TABLE . . . AS SELECT statements, which select from external tables, require the APPEND
hint to use direct path loading.
D. INSERT INTO . . . SELECT FROM statements, executed serially, which select from external tables,
are unable to use direct path loading.
A CTAS (Create table as select) will always use direct path (B, not C) load but IAS (Insert as select) statement will not. In order to achieve direct path load with an IAS statement you must add the APPEND hint to the command (A, not D).
Direct path loads can also run in parallel. You can set the parallel degree for a direct path load either by adding the PARALLEL hint to the CTAS or IAS statement or by setting the PARALLEL clause on both the external table and the table into which the data will be loaded.
Once the parallel degree has been set at CTAS will automatically do direct path load in parallel but an IAS will not. In order to enable an IAS to do direct path load in parallel you must alter the session to enable parallel DML.
* Parallel Direct Path Load
The key to good load performance is to use direct path loads wherever possible. A direct path load parses the input data according to the description given in the external table definition, converts the data for each input field to its corresponding Oracle data type, then builds a column array structure for the data. These column array structures are used to format Oracle data blocks and build index keys. The newly formatted database blocks are then written directly to the database, bypassing the standard SQL processing engine and the database buffer cache. Reference: Best Practices for Implementing a Data Warehouse on the Oracle Exadata Database Machine; Using CTAS & Exchange Partition Replace IAS for Copying Partition on Exadata