What's New

This section summarizes the new features of Oracle In-Memory Database Cache release 11.2.2 that are documented in this guide and provides links to more information.

New features in Release 11.2.2.4.0

• New instructions have been added on how to backup and restore a TimesTen database that contains one or more cache groups. For more details, see "Backing up and restoring a database with cache groups".

• A new tool, the Cache Advisor, has been added to help determine whether the performance of an existing Oracle Database application that runs a workload of SQL statements can be improved when the application is used with a TimesTen database. Cache Advisor analyzes application performance and generates recommendations of TimesTen cache group definitions based on the SQL usage in the Oracle Database application. For more information, see Chapter 9, "Using the Cache Advisor".

New features in Release 11.2.2.2.0

• You can configure parallel propagation of changes in AWT cache tables to the corresponding Oracle tables using either the ReplicationParallelism or CacheAwtParallelism data store attributes. See "Configuring parallel propagation to Oracle tables".

• If you are using parallel propagation, any unique index, unique constraint, or foreign key constraint on the columns of the cached Oracle tables must also be created on the cached tables in the AWT cache group. See "Configuring parallel propagation to Oracle tables".

New features in Release 11.2.2.1.0

• You can configure parallel propagation of changes in AWT cache tables to the corresponding Oracle tables. See "Configuring parallel propagation to Oracle tables".

• The default value for the cacheAWTMethod first connection attribute has changed. See "Improving AWT throughput".

New features in Release 11.2.2.0.0

• You can obtain information about the grid node where a global query is being executed. See "Obtaining information about the location of data in the cache grid".

• You can perform a local join when executing a global query. See "Performing global queries with local joins".