Identify two reasons why using external tables for high performance data loads on Exadata is
more advantageous than SQL*Loader.
A.
Transformations can be applied directly on the file data using SQL or PL/SQL constructions.
B.
External tables allow transparent parallelization inside the database.
C.
Parallelizing loads with external tables enables more efficient space management.
D.
External tables can take advantage of storage indexes to speed up data loads.
Explanation:
Parallelizing loads with external tables enables a more efficient space management
compared to SQL*Loader, where each individual parallel loader is an independent
database sessions with its own transaction.
Answer is A,C
A, C:
data can be transformed using sql or pl/sql and the loads can be parallelized.
It’s B & C – see Page 6: http://www.oracle.com/technetwork/database/bi-datawarehousing/twp-dw-best-practices-for-implem-192694.pdf
If you are loading from files into Oracle you have two options, SQL*Loader or external tables. Oracle strongly recommends that you load using external tables rather than SQL*Loader:
• Unlike SQL*Loader, external tables allows transparent parallelization inside the database.
• You can avoid staging data and apply transformations directly on the file data using arbitrary
SQL or PL/SQL constructs when accessing external tables. SQL Loader requires you to load
the data as-is into the database first.
• Parallelizing loads with external tables enables a more efficient space management compared
to SQL*Loader, where each individual parallel loader is an independent database sessions with
its own transaction. For highly partitioned tables this could potentially lead to a lot of wasted
space.
As per the OraDBA doc Ans is A,B & C.
Latest exam materials http://www.exadata-certification.com
A B C
A,C
From the guide:
Benefits over SQL*Loader:
– Transformations can be applied directly on the file data using SQL and PL/SQL
– Parallelizing loads with external tables enables efficient space management
B is not wrong since it breaks up big files into 10MB granules, but the question states: Identify two reasons