The first step in any archival process is to identify what determines stale or inactive data. This is often a matter of subjective interpretation of business rules and is therefore best performed by the business lines whose data is under scrutiny.
The next step is to define the collection of records, perhaps spread across multiple databases and files, which relate to the inactive data. Data Archiving for Adabas describes such a collection of records as a business object.
Once such business objects can be identified, the next step is to choose the most efficient means of selecting these objects. Data Archiving for Adabas offers three different data selection methods:
This method uses a simple scripting language to allow you to select business objects and to perform the archiving/transfer actions to be taken on those objects.
For detailed information refer to the ADRExtractionSyntaxReference.pdf document located in the product's sdk/doc directory.
Some simple introductory examples on how to use the syntax can be seen here Extraction Syntax Examples.
This method allows you to select business objects with your own custom applications (written in Natural or C programming language) using the Data Archiving for Adabas Application Programming Interface . The application programming interface enables the ability to ‘stage’ selected business objects in a repository file, an Action is then run which references these staged business objects and performs the appropriate archiving /transfer operations. Refer to Data Selection Mode for information on how to define an Action to run in To-Do list mode.
This method allows you to create a C programming language plug-in which, in conjunction with the Data Archiving for Adabas Application Programming Interface, provides business objects at runtime when requested by the Data Archiving Service Extractor process. Refer to Data Selection Mode for information on how to define an Action to run in User Library mode.