Create A Time-Based Partitioned Cube Group; Processing Issues Related To Time-Based Partitioned Cubes - IBM Cognos User Manual

Create a Time-based Partitioned Cube Group

To more easily manage time-based partitioned cubes, you can gather them into
like-structured groups.
Procedure
1. To set up the cube group, on the PowerCube property sheet, click the Cube
2. To ensure that the child cubes in your group cover a distinct level, for each one,

Processing Issues Related to Time-based Partitioned Cubes

Time-based partitioning can change how categories are sorted, depending on
which categories are included in a particular partition.
Every child category does not appear in every child cube. Category inclusion is
based on the time level specified when you enable partitioning. Data outside the
specified time period is not included in the time-based partitioned cube.
Your goal as modeler is to define the partitioning so that every child cube contains
all the relevant or possible categories; that is, an inclusive approach. However, this
may not be feasible if, for example, all new categories are added only to the latest
cube. In such cases, the sort order for the categories is stored in the individual
child cubes, called disjoint cubes, and not in the large cube.
Sorting Considerations
Even when the control cube contains high-level metadata about the overall
structure of the cubes, frequent changes to the metadata may affect how
time-based partitioned cubes perform. For sorting to work properly, the changes to
the categories must be relatively slow and orderly, and there must be no significant
changes to the metadata categories in the overall model.
If you make any of the following changes to your model, unexpected sorting may
result:
v multiple category deletions
v the addition of many new categories that lack contextual information about their
v use of the Unique and Move options for categories in a level
Because time-based partitioned cubes handle large amounts of time-related OLAP
data in a seamless, efficient manner, cube updates and query times are normally
faster. However, querying may be slower due to the extra time needed to sort
across multiple child cubes, and there may also be an increased memory
requirement. This performance overhead varies, depending on such factors as the
number of cubes referenced by the query and the number of child categories in
each parent being sorted.
128
IBM Cognos Transformer Version 10.1.1: User Guide
Group tab, select the Enable time-based partitioning check box, and click OK.
specify the appropriate level from your time dimension, such as Quarter or
Month.
Tip: You can open the .vcd file later, in any text editor, and manually include
or exclude cubes. For example, to improve performance, try adding entries in
the .vcd file for cubes that are at a higher level of the time hierarchy, such as
the Quarter or Year level. For more information, see "Customizing a
Time-based Partitioned Cube" on page 131.
parent-child and child-sibling relationships