What is the Archive

In the context of Valuemation archiving, an archive is a database (or 'database schema' in Oracle terminology) of the same structure as the main Valuemation database but with the following differences:

• The archive schema initially contains no data

  The archive schema is created as a copy of the database structure but with no data. The only way to fill the archive with data is via the archiving process. This also means that care must be taken not to import data into the archive during possible structural updates due to new functionality or hotfixes. See 'Archive Maintenance'.

• Foreign key constraints are omitted

  Foreign key constraints present in the main database are not copied during the archive creation. The absence of foreign key constraints enables related data to be stored in two different database schemas. For example, the following can be expected for archiving of Tickets and their related object types:

  • Tickets - archived
  • Ticket Descriptions - archived
  • Ticket Classes - not archived

  The task is to archive "stale" tickets (root object type) and objects referencing these tickets (e.g. ticket descriptions). Objects referenced by the tickets (e.g. ticket classes) do not have to be (and usually are not) archived. In the case of ticket classes, ticket classes are classification objects not expected to reach large quantities and expected to be continually used for classification of new objects in the main schema. For these reasons ticket classes stay in the main schema and are referenced by the archived tickets from the archive.

• Indexes are not archived

  Indexes (with the exception of primary key column indexes) are omitted for performance reasons. The presence of indexes would exponentially slow down the initial stage of the archiving process (i.e. filling the newly created archive schema with the first bulk of archived data).

  The absence of indexes in the archive will slow searches on archived data by a certain extent. Should the need for viewing large quantities of archived data arise, the archive can be re-indexed after the initial archiving action.

• Database Triggers are omitted

  Database triggers usually react to the deletion, creation or updating of data. This is unwanted behaviour in the archive because archived data are not meant to be further modified.