IBM Cognos, User Manual

Page: 140 / 411

v
maintains a list of all measures, dimensions and root categories, currency
records, and the entire structure of the time dimension
v
contains the control information to gather the data from the child cubes
v
serves as the entry point for OLAP report users to access the time-based
partitioned cube
The .mdc file shares the same file name and location as the .vcd file.
For example, suppose your sales organization can benefit from a time-based
partitioned cube group defined on the Quarter level, so cubes can be rebuilt every
quarter.
After one year, you have separate cubes for the four quarters of 2006. In the first
quarter of 2007, you run an update creating a 2007_Q1 child cube. The large
control cube has categories in the time dimension for 2006 and 2007. The text file
(.vcd) generates five lines, one for every quarter from 2006_Q1 to 2007_Q1.
Advantages of a Time-based Partitioned Cube
Time-based partitioned cubes are similar, but superior, to standard cube groups, in
that they
v offer a faster, more efficient way of building and updating time-segmented data
than incrementally updated cubes
New data is typically added to a single partitioned cube, rather than to a large
existing cube.
v
eliminate the periodic need to do a full build, resulting in shorter down-times
for the production system, and a more easily managed maintenance schedule
v
support rolling time periods
You can manually edit the .vcd file to remove references to cubes that are no
longer required, and drop invalid or out-of-date categories. The control cube and
definition file are automatically updated with the newest categories and cube
references. For more information, see “Customizing a Time-based Partitioned
Cube” on page 131.
v support slowly changing dimensions
Existing child cubes retain the history, and new cubes are easily created using
the
Move capability for categories.
v
offer better query performance, because users drilling down into the time
dimension encounter fewer cubes
When report users reach the level of granularity that the cubes are based on,
such as January or Q1, they only need to access a single cube.
v
offer more flexibility
Although time-based partitioned cubes relate to only one level of granularity,
such as Month, you can still reference other time levels in the same model or
cube if this improves run-time performance.
Disadvantages of a Time-based Partitioned Cube
Category handling, sorting enhancements, and support for external rollup make
time-based partitioning a useful optimization technique for production
environments.
However, the following restrictions apply:
126
IBM Cognos Transformer Version 10.1.1: User Guide