+

Search Tips   |   Advanced Search

Create scheduler databases

The scheduler uses the scheduler database for storing and running tasks. To create a scheduler database, the database system must be installed and available.

The performance of schedulers is ultimately limited by the performance of the database. If we need more tasks per second, we can run the scheduler daemons on larger systems or we can use clusters for the session beans used by the tasks. Eventually, however, the task database becomes saturated and you then need a larger or better-tuned database system.

Multiple applications can share a scheduler database. This sharing can lower the cost of administering scheduler databases.

The scheduler requires a database, a JDBC provider, and a data source.

  1. Create the database according to the description for your database system:

    • Create Apache Derby databases for schedulers.

    • Create a DB2 database for schedulers.

    • Create a DB2 database for z/OS for schedulers.

    • Create a DB2 for iSeries database for schedulers.

    • Create an Informix database for schedulers.

    • Create a Microsoft SQL Server database for schedulers.

    • Create an Oracle database for schedulers.

    • Create a Sybase database for schedulers.

  2. If the database is not on the same machine as the IBM WebSphere Application Server, verify that we can access the database from the application server machine.

  3. Configure the JDBC provider and data source. For details, refer to the Create a JDBC provider and data source topic. The JDBC driver can be either one-phase or two-phase commit depending on whether other transactions take place using other data sources, for example, while using the scheduler. The data source can represent multiple versions of the product.


Results

The database is created and ready for you to create scheduler tables.


Related tasks

  • Create scheduler tables using DDL files
  • Configure a JDBC provider and data source