Batch mode configures clients to disconnect from the middle tier. With the client application running in batch mode, requests are bundled together for later processing. Browsing, updating, deleting, etc., are all done on the client without going over the network. When the transaction is committed, the changes are bundled together and sent. This greatly reduces the number of network roundtrips, improving scalability and performance.
Batch mode also gives JClient applications the ability to use application module pooling for the first time.
Note: Batch mode is not supported for applications developed in JDeveloper 9.0x.
Clients created in JDeveloper are set to batch mode by default. This may not be appropriate for applications that have a very high level of interactivity with the database and/or where immediate validation is required. In this case the application should be run in immediate mode so that data is synchronized immediately.
Batch mode is set in the Property Inspector for the Data Control (.cpx). To change the setting see Changing Query Data Synchronization.
Batch mode and immediate mode behave slightly differently and may require different coding strategies. Also, if you have a legacy application that you want to change to use batch mode you will need to make some changes in the code. For more information, see the topic Coding Considerations in Batch Mode.
Changing Query Data Synchronization
Copyright © 1997, 2004, Oracle. All rights reserved.