140 IBM Cognos Dynamic Cubes
For example, assume that you have a member with a business key of 1066 in one cube and a
business key of 1415 in another. You want to merge them because they represent the same
entity.
Defining virtual measures
Measures that do not have identical names or that exist in only one of the source cubes are
added to a virtual cube as new virtual measures.
If a virtual measure is not merged correctly, or cannot be automatically merged, you may
manually merge two source measures. You may also delete redundant virtual measures.
When merging measures in a virtual cube, it is not possible to map a source measure to more
than one virtual measure.
6.3 Common use cases for virtual cubes
This section describes several scenarios of when virtual cubes might be useful for your
application.
6.3.1 Time partitioning
In this scenario, all sales information was originally stored in one cube named AllSales. Fact
data is added nightly, which requires rebuilding the data cache and refreshing the summary
tables. For large cubes, this task requires a significant amount of time. To improve
performance, you can partition the data for the AllSales cube based on time to form the basis
for two cubes: HistoricSales and CurrentQuarterSales. And, you may define a virtual cube,
named VirtualSales, to join the two cubes. The HistoricSales cube is used to record the
historical sales information that does not change frequently. The smaller CurrentQuarterSales
cube is used to record the daily sales information for the current quarter. The technique is
referred to as the
delta cube.
6.3.2 Currency conversion
Assuming that all sales information from the company is stored in one cube named
GlobalSales, and there is a business need to convert some of the sales figures that are in US
dollars into other currencies. If both the sales and exchange rate information are in the
GlobalSales cube, the cube contains much redundant data and is difficult to maintain. To
address this problem, you can create a cube named CurrencyExchange to store the
exchange rates to keep the Global Sales cube unchanged, and you can then define a virtual
cube named SalesConversion to handle the currency conversion for the sales data. The
GlobalSales and CurrencyExchange cubes share some dimensions, such as Time, but they
generally have different structures.
The SalesConversion virtual cube needs to define a calculated member that handles the
actual currency conversion. Because the actual currency conversion might not occur daily but
instead only at preset days or at intervals such as monthly or weekly, the ClosingPeriod and
OpeningPeriod functions can be used in this scenario.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.