Multiple geodatabases in DB2

How multiple geodatabases are stored in a DB2 instance varies depending on whether you are using IBM DB2 for Linux, UNIX, and Windows platforms or DB2 for the Z operating system (z/OS). The following sections explain how they are stored on different platforms.

DB2 for Linux, UNIX, and Windows

For DB2 on Linux, UNIX, or Windows operating systems, you store one geodatabase per spatially-enabled database.

If you need to have multiple geodatabases from the same DB2 instance on the same server machine, you should set up your first geodatabase and its ArcSDE service (if you are using a service), then create additional DB2 spatially-enabled databases and create a geodatabase in each.

If you only use direct connections to the separate geodatabases, no special configuration is required; when you connect with a direct connection, the database name or alias differentiates to which geodatabase you are connecting. However, if you use an ArcSDE service to connect to the geodatabases, you must have separate ArcSDE services with unique service names and a separate port number in the services file for each geodatabase.

DB2 for z/OS

For individual ArcSDE geodatabases stored in DB2 for z/OS, at least two databases are used in the database subsystem—one to store the ArcSDE geodatabase repository and at least one to store user data. This is illustrated in the following graphic:

To set up multiple geodatabases, you need multiple database subsystems.

Tip:

See your DB2 z/OS documentation for further information on setting up a database subsystem.

Since connections to geodatabases on DB2 for z/OS are only made using direct connections, you just need to be sure to specify the correct database subsystem name when making a connection from the client machine, including when issuing ArcSDE administration commands.