multiple-databases
Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
multiple-databases [2016/01/22 15:33] dmtolpeko [Working with Multiple Databases in PL/HQL] |
multiple-databases [2016/01/22 15:37] dmtolpeko [Map Database Objects to Use Other Connections] |
||
|---|---|---|---|
| Line 7: | Line 7: | ||
| Hadoop extends, not replaces a traditional data warehouse, so you have to work with different systems for different type of workloads. | Hadoop extends, not replaces a traditional data warehouse, so you have to work with different systems for different type of workloads. | ||
| - | Most SQL-on-Hadoop solutions are good for querying large volumes of data rather than transaction processing. Note that Hive stores its metadata is a separate database (MySQL, PostgreSQL, Oracle etc.). | + | Most SQL-on-Hadoop solutions are good for querying large volumes of data rather than transaction processing. Note that even Hive stores its own metadata is a separate database (MySQL, PostgreSQL, Oracle etc.). |
| When you use Hive or other SQL-on-Hadoop tool to query your data, you can use a RDBMS or NoSQL to: | When you use Hive or other SQL-on-Hadoop tool to query your data, you can use a RDBMS or NoSQL to: | ||
| Line 20: | Line 20: | ||
| Before you can work with multiple databases, you have to [[connections|configure connections]]. | Before you can work with multiple databases, you have to [[connections|configure connections]]. | ||
| - | PL/HQL connects to a database only when it executes the first SQL statement for this database, the tool does not connect to any database in advance. | + | HPL/SQL connects to a database only when it executes the first SQL statement for this database, the tool does not connect to any database in advance. |
| - | Once connected, PL/HQL holds and re-uses the connection until the script completes. | + | Once connected, HPL/SQL holds and re-uses the connection until the script completes. |
| ===== Default Connection ===== | ===== Default Connection ===== | ||
| - | By default, PL/HQL uses the connection profile defined by the [[configuration#plhqlconndefault|plhql.conn.default]] option. | + | By default, HPL/SQL uses the connection profile defined by the [[configuration#hplsqlconndefault|hplsql.conn.default]] option. |
| ===== Map Database Objects to Use Other Connections ===== | ===== Map Database Objects to Use Other Connections ===== | ||
| - | Once you configured multiple connection profiles you can use [[map-object|MAP OBJECT]] statement to map a table or view to the specified connection. Then when accessing in SQL statements PL/HQL will use appropriate connection for the object: | + | Once you configured multiple connection profiles you can use [[map-object|MAP OBJECT]] statement to map a table or view to the specified connection. Then when accessing in SQL statements HPL/SQL will use appropriate connection for the object: |
| For example, you can run SELECT in Hive, but log messages to MySQL: | For example, you can run SELECT in Hive, but log messages to MySQL: | ||
Page Tools
