Archiving Object 

Definition

A logical object containing related business data in a database that can be read from the database with a write program, and, after successfully being archived, can be deleted from the database using a corresponding delete program.

Use

To archive data from InfoProviders, you first need to create an archiving object for the InfoProvider. InfoCubes and DataStore objects usually contain a self-enclosed dataset for a specific business-related area. For this reason, an individual archiving object is generated for each InfoProvider. Only one archiving object can be created for each InfoProvider.

First of all, maintain the metadata of the archiving object in the InfoProvider The actual object is not generated until the InfoProvider is activated.

You can also create the archiving object afterwards for an InfoProvider that already exists and is filled with data.

Structure

The following steps are required to define an archiving object:

Select Data

Time slot

You can enter a time here when the system is to carry out archiving. If you choose None, you need to choose selection characteristics in the lower half of the screen instead. By choosing By Characteristic, you can use the dropdown list box to select just a time characteristic or a characteristic that references a time characteristic.

Choose this time slot mode if you want to archive regularly. In this way, the selection screen of the write program changes also, whereby you reduce the administration workload, because a new variant does not need to be created for each archiving run. In the maintenance for variants, you can specify whether you want a relative and/or absolute time restriction. See also: Time Restrictions

Only for DataStore objects: If the Protect Archived Area indicator is set, the system checks new data before it is activated to see if these data records that are in a time slot have already been archived. If this is the case, the system terminates activation and writes the affected archiving runs to the archiving log. For additional procedures you have two alternatives:

·        To activate the data successfully, you can reload the archiving runs, which were specified in the log, into the table of active data for the DataStore object by using the archiving administration (see Reloading Archives). Activation can now be started again.

·        Or delete the affected requests from the activation queue for the DataStore object, modify the update rules so that the affected data records are skipped, and load and reactivate the requests.

The protection of archived data areas is also possible with a time characteristic, which is not a key characteristic of the DataStore object. In some cases, however, you might need to check whether the protection is worthwhile or not. For example, protection cannot be ensured if the value of the time characteristic can be subsequently enlarged in a data record (and, in doing so, lies outside of the protected area). A time characteristic in the data portion that is derived from the archiving status in the source system can, however, be by all means useful.

This key figure needs to be set for archive administration to be able to allow data to be reloaded directly into the active data table for a DataStore object.

 

Choose selection characteristics

The characteristic you have specified for time selection is displayed here. You cannot choose any additional selection characteristics.

If you have not specified a characteristic for time selection, you must set selection characteristics here. One of these characteristics must be a key characteristic and must display a measurement for the age of the data. The following criteria are available for the selection:

·        If the InfoProvider is partitioned, the selection characteristic must be identical to the partition characteristic.

·        If the InfoProvider is not partitioned, you must select a time characteristic from the key.

If neither of these conditions is fulfilled (this is possible for DataStore objects, for example), additional characteristics of the InfoProvider can be used as selection characteristics for archiving sessions. Examples are the characteristics such as product (when a product is being phased out, for example) or company code/company (when an organization sells or unloads a legal unit).

The archiving method cannot be changed after the first activation run. For more information, see Changing or Deleting an Archiving Object.

File Structure

Define a maximum size for the generated archive files. You must take the memory capacity of your storage medium into account. The maximum size is specified in Kbytes. It can be restricted by the number of data objects (see below). The limit to be reached first is the determining one. During an archiving run, a new file is created when the maximum size is reached.

Define your data object in the lower half of the screen. A data object is the smallest indexable and accessible unit in the archive. A data object for transaction data can be created using a quantity of records that have the same values (= grouping characteristics) with respect to the key characteristic selection. The system reads from the database after sorting according to grouping characteristics (the order is important). Records with the same specifications in the grouping characteristics are written to the archive as data objects. In some circumstances, a data object defined in this way can be subdivided again into several data objects to limit size. The grouping characteristics thereby form the basis of a possible archive indexing.

The data is then sorted according to these characteristics and the same records are grouped in a data object. If no characteristics are selected here, data is written to the archive file as unsorted.

Store

Here you must specify the logical file name, storage system and delete sequence. You only need to specify the storage system and delete sequence if you are using a separate storage system.

Generating an Archiving Object

An archiving object is generated locally from the archive-specific attributes when activation takes place.

The name range for archiving objects consists of: ‘BW’ as the prefix, then a character for the object type (‘C’ = InfoCube, ‘O’ = DataStore object A version), then comes the object name, for which only seven characters are available, since only ten characters are available for the whole name.

A mapping table is created. This assigns the BI InfoProvider an unambiguous name for an archiving object.

When the InfoProvider is deleted, the archiving object and the mapping table entry are also deleted. In addition to the archiving object, programs to write to the archive and to delete from Archive Administration are generated.

Integration

Deleting an InfoProvider also deletes the archiving object assigned to it. This also deletes links to the archive files. A confirmation prompt appears before you delete the InfoProvider if the archive contains data.

When deleting the content of an InfoProvider, only data from the database tables is deleted. Links to the archive files remain in place.

See also:

Change or Delete an Archiving Object

Archiving Object (SAP NetWeaver)