Standard SUM versus SUM with DMO
Consider the following main differences to the standard SUM procedure, that is, the system update without combined database migration in a one-step procedure:
-
Different prerequisites
Supported are all database versions listed in the Product Availability Matrix (PAM) of the source release and all operating system versions listed in the PAM of the target release.
-
Download directory
One of the first DMO dialogs in the browser is the question where the stack configuration file stack.xml is located. SUM presumes that this file is part of the download directory, so there is no separate question about the location of the download directory. The download directory includes the kernel and software files that were downloaded after using the Maintenance Planner.
-
Additional Steps
The database migration is an additional part of the DMO procedure compared to the standard update procedure of the SUM. Therefore, you must provide specific information for the migration and for the target database during the DMO procedure, such as a migration key.
Technically, SUM creates the shadow repository on the target database and triggers the R3load to export and import the application data.
-
Easy reset
An easy reset to the source database is possible, as long as the source database and the SUM directory (of the existing update) exist. Do not perform a reset in a productive environment after the users started working on the new system based on the new database.
-
Declustering of cluster tables
Because of efficient compression algorithms of IBM Db2 for Linux, Unix, and Windows, SAP cluster and pool tables offer no advantages. Therefore, during the migration from the source to the target database IBM Db2 for Linux, Unix, and Windows, all cluster tables are automatically de-clustered. The DMO also decompresses all pool tables. The target system on IBM Db2 for Linux, Unix, and Windows contains only transparent tables.
