Show TOC

Editing Aggregates ManuallyLocate this document in the navigation structure


The InfoCube for which you want to create an aggregate has been saved and is active.


How you manually create or change an aggregate for an InfoCube is described below.

Access from Data Warehousing Workbench

  1. You are in the Data Warehousing Workbench in the Modeling functional area.In the navigation window, choose InfoProvider and in the InfoProvider tree, navigate to the InfoCube whose queries you want to optimize.
  2. In the context menu of the InfoCube, choose Maintain Aggregates.The Maintain Aggregates screen appears. If an aggregate has already been created for the selected InfoCube, you can also get to the maintenance screen by double-clicking on .

Using Transaction RSDDV

  1. On the Aggregate/BW Accelerator Index: Select InfoCube screen (transaction RSDDV), select the required InfoCube.
  2. Choose Aggregates. The Maintain Aggregates screen appears.

    If you are creating the first aggregate for an InfoCube, the Proposals for Aggregates dialog box appears first. You can choose whether the system proposes aggregates or whether you want to create them manually.

    For more information, seeCreating the First Aggregate for an InfoCube.

  3. The left side of the screen shows the dimensions, characteristics and navigation attributes of the selected InfoCube in a tree structure as Selection Options for Aggregates.

    Select one or more objects to be copied to the aggregate.


    Define the granularity you require for the data in the aggregate. Add all the characteristics derived from these characteristics.


    For example, if you define an aggregate for the month, you should also include the quarter and year in the aggregate.

    This enhancement does not enlarge the dataset, but allows:

    • A year aggregate to be built from this aggregate
    • Those who need the annual values to use the queries for this aggregate.

      You can only include a characteristic and one of its attributes in an aggregate in expert mode (Extras → Switch Expert Mode On/Off). An aggregate of this type has the same granularity and size as an aggregate that has only been built using the characteristic, but is affected by the hierarchy/attribute change run. Compared with the aggregate for the characteristic in which the attribute information is defined by a join with the master data table, the aggregate for the characteristic and the attribute only saves the database join.

      We therefore recommend that you either build an aggregate using the characteristic or you build a much smaller aggregate using the attribute.

  4. You have different options for creating an aggregate:

    Transfer the selected object(s) to the Aggregates column on the right side of the screen using drag and drop.

    Select Create New Aggregate.

    The Enter Description for Aggregate dialog box appears.

  5. Enter:

    Short description

    Long description


    To change the text at a later time, select Change Description Text in the context menu of the aggregate.

  6. Choose Continue. The Maintain Aggregates screen appears.

    The system displays the aggregate in the top-right area of the screen. The log is displayed in the lower part of the screen.

    For more information, seeDisplaying an Aggregate and Its Components.

  7. If an aggregate contains a time-dependent component, you must assign the aggregate a key date.

    When you fill the aggregate, the key date behaves like the key date of a query: The time-dependent attributes and hierarchies are evaluated on this key date. For this reason, aggregates with a time-dependent component can only be used in a query if the key date of the query is the same as the key date of the aggregate.

    In the Select Variable or Fixed Date dialog box, select the following as the key date:

    A variable that is also used in queries for the key date and can be automatically calculated in the SAP Exit or Customer Exit processing types (seeVariables),


    0CWD current workday

    0DAT current calendar day

    0P_KEYDT key date of due date

    0P_KEYD2 key date of posting (from key date of due date)

    0P_KEYD3 key date of clearing (from key date of due date)

    0P_KEYD4 key date of posting (posting date)

    0P_KEYD5 key date of clearing (from key date of posting)


    Aggregates with a variable key date must be updated regularly. This process must be included in aProcess Chain (Further BW processes -> Adjust Time-Dependent Aggregates).

    a particular calendar day


    To enter a calendar day, select object CALENDAR (<Calendar>) in the Select Variable or Fixed Date dialog box. Choose Transfer Selections. The Calendar dialog box appears. You can copy this date to the aggregate definition by double-clicking on it.

    In the aggregate tree, under Properties for node Variables for Key Date, the system displays the technical name of the variable you chose or the fixed calendar date.

    Once the aggregate has been activated and filled, the system copies the key date computed from the variable for when the aggregate should be filled, into line Key Date.

  8. By default, the system partitions the aggregate fact tables when the related InfoCube is partitioned and the partitioning characteristic is available in the aggregate. Choose Properties → Change Partitioning to prevent partitioning for individual aggregates.

If aggregates do not contain much data, very small partitions can result. This affects read performance. Aggregates with very little data should not be partitioned. Note that if you change this property to Not Partitioned for an existing aggregate, you have to activate and fill the aggregate again.

  1. You can change the structure of the aggregate by adding additional components or deleting existing ones. You can also change the key date.

    Inserting components into the aggregate

    Select one or more objects in Selection Options for Aggregates.

    Use drag and drop to transfer them to the aggregate that you want to change on the right-hand of the screen.

    Where necessary, change theselection type (BW-WHM) by choosing the appropriate entry in the context menu:

All characteristic values

Hierarchy level

Fixed value


Aggregates containing fewer than 14 components are stored on the database in optimized form (seeEfficiently Loading Data into Aggregates).

Note that the characteristics that are defined in the InfoCube are also included in the aggregate and thus increase the number of components, even though they are not visible on the interface.

Deleting components from the aggregate

In the aggregate tree, navigate to the characteristic(s) or navigation attributes that you want to delete.

Select these components.

To delete the components from the aggregate, you can:

In the context menu, choose Remove Components.

Use drag and drop to copy the components to the left part of the screen.


To delete a dimension from an aggregate, you have to delete all the characteristics and navigation attributes of this dimension.

Changing the key date

In the aggregate tree under Properties, select the node Variable for Key Date and choose Change in the context menu. The Select Variable or Fixed Date dialog box appears.

Select and transfer the required variable or calendar day.


The key date computed from a changed variable is only copied to line Key Date if the Adjust Time-Dependent Aggregates process has been executed.

  1. To check the aggregate definition for inconsistencies, choose Check Definition.
  2. Save the new or changed aggregate.

You can activate the new or changed aggregate and fill it with data. It is available for reporting.