Table of Contents Previous Next


4 Using the Hadoop Data Adapter

When possible, the data adapter pushes down the WHERE predicate and target column list to the Hive server. Supporting operations (aggregates, foreign joins, and sorts) are performed by the database server. Remote push down provides better performance and improves network traffic by distributing processing to the remote server when possible, decreasing the amount of data that must travel to the database server.
You can demonstrate command push-down by including the EXPLAIN clause in a query of a foreign table:
As noted near the end of the query plan, the WHERE clause is performed on the Remote server.

4 Using the Hadoop Data Adapter

Table of Contents Previous Next