The high-level CMA process comprises the following tasks and flows. Each is described in more detail later in this section.
The primary instruction in the migration plan could define the physical BO of the record. This signals to CMA that all records for the MO are eligible to be migrated.
The primary instruction may define a specific BO, in which case only records that reference that BO in its parental hierarchy are considered.
Subsequent instructions may include references to related records. There may be some circumstances where a migration should include subsequent records and some circumstances where they shouldn’t based on requirements.
A Migration Request is used to define a set of migration plans to be included together in a single migration attempt. In addition, the migration request may define selection criteria for each migration plan to limit the migration to a subset of data for each MO. Selection may be defined using SQL, an algorithm or explicit primary key values.
A Migration Data Set Export object is created for a specific migration request.
The lifecycle of the Migration Data Set Export business object includes algorithms that select the appropriate records according to the migration request, determine dependencies between records to build groupings of related objects and create the export file.
Because there may be a large volume of data in the export, many of the steps in the lifecycle of the migration data set export are executed via the Migration Data Set Monitor.
The file created by the export, which is a BINARY file, needs to be transferred from the export directory to the import directory. The transfer needs to be done in such a way as to preserve the file structure. Refer to Migration Assumptions, Restrictions and Recommendations for more information.
A Migration Data Set Import object is created for a given file to import.
The next step reads the import file and creates Migration Transactions and Migration Objects based on the information in the import file. A migration object is created for every maintenance object record to potentially be created or updated. The migration transaction is a grouping record that groups together related migration objects.
The next step is for the objects to Compare the data being imported against the data for that record in the target environment. If it is found that the migration object is new or represents a change to the existing record in the target environment, appropriate SQLs are generated. If no changes are found, the object is marked “unchanged” and doesn’t progress.
Once all the objects are compared, the user may review the objects for acceptance or rejection.
When the migration objects are all accepted or rejected, the next step is to apply the objects and update the target environment.
Copyright © 2007, 2016, Oracle and/or its affiliates. All rights reserved. Documentation build: 2.5.2016 10:21:45 [T1_1454696505000]