Rolling up Data into an Aggregate 

Use

If you load new data packages (requests) into the InfoCube, these are not available at first in reporting by using an aggregate. In order to supply the aggregate with the new InfoCube data, you first have to load these into the aggregate tables for a set time frame. This process is known as a roll up.

Prerequisites

Procedure

In the InfoCube maintenance you can set for each InfoCube the way in which the rolling up of data packages into the aggregates need to be run.

You are in the Modeling function area in the Administrator Workbench. Choose Manage by using the context menu for the InfoCube you want. You arrive at the screen Managing the Data Targets. The system transfers the InfoCube data into the list in the upper screen area.

Execution Type

Procedure

Including data package into a process chain as a process

We recommend admission into a process chain
      • with complex schedules
      • with difficulties with automatisms and event collectors
      • for all new developments
  1. Choose the tabstrip Roll Up from the screen Managing Data Targets.

The system suggests the largest possible value for Roll Up to Request ID. However, you can overwrite this.

  1. Choose Process Chain Maintenance.

 

By using the transaction RSPC, you arrive at the dialog box Selecting a Process Chain with an overview using the existing process chains in the BW system.

If there is no suitable process chain, you can create a new process chain with the process of rolling up.

You can find additional information under Creating Process Chains in the section Creating Process Chains by Using a Maintenance Dialog for a Process.

Starting data package roll ups manually

 

Use this procedure especially when the data from several data packages create a logical unit, and they therefore only need to be released.

Different plants deliver their data at different points in time. The data needs to be visible first, however, in the InfoCube if all plants have loaded their data into the InfoCube.

  1. Choose the tabstrip Roll Up from the screen Managing Data Targets.
  2. Choose Selection. You get to the Start Date Values dialog box.
  3. Choose the start date. You have the following options:
    • Immediately
    • Date/Time
    • After job
    • After event
    • By operation mode
  1. Set the respective indicators if you would like the job to run periodically.
  2. Save

The following procedures are also possible, but should not be used for new scenarios.

Execution Type

Procedure

Subsequent processing
  1. Choose the tabstrip Roll Up from the screen Managing Data Targets.
  2. Choose Subsequent Processing. You arrive at the dialog box Maintain Events to Run After Processing.

You can find further information under Process Control and Creating an Event Collector.

Rolling up each data package automatically into the aggregate

 

The InfoCube has to be ensured for technical correctness and quality.

Only use the automatic roll up when you load requests into the InfoCube and there is no overlapping between processes for loading, for roll ups, and for other automatisms for the InfoCube.

You can find additional information under Automatic Further Processing.
  1. Choose this from the screen Data Target Administration from the Environment -> Automatic Request Process You get to the Automatism Maintenance dialog box.
  2. Set the indicator for Roll Up Data into the Aggregate under the group heading Automatic Processing.
  3. Save.

Program RSDDK_AGGREGATES_ROLLUP

You can also run the roll up with the program RSDDK_AGGREGATES_ROLLUP (see Running Programs Directly - Reports).

You can also schedule this program as a regular background job or use it in an event collector.

Result

The new data for these queries, that were started after the roll up, is taken into account in reporting.

See also:

Managing InfoCubes