Databases Reference
In-Depth Information
The business integrity of the data is guaranteed thanks to
the control of lifecycles at a business level. Even if the data
update approval process malfunctions, the data in the
system retains its integrity as long as the business lifecycles
are respected.
If the concerns of the business and the organization are
not separated, we run the risk of creating an MDM system
with limited flexibility, even harmful in regards to the
integrity of the data. In that case, an evolution in the
organization would require an update which would have an
impact on the MDM system at a business level even though
it evolves following a different lifecycle.
8.1.1.2.
Managing data integrity
To separate the business from the organization, it is
necessary to identify the business state of each data element
used over the course of time. In practice, this state is
attributed to an aggregation of data grouped under a
business object. In this way, according to the various states
of the business object, updating is or is not possible. Certain
updates will always be authorized, irrespective of the
business object's states. By distinguishing the operations
which are dependent on the business objects' states from
those which are not dependent, the modeling is more
rigorous, clear and protective of the data integrity.
If this distinction is not taken into account in the
modeling, then we run the risk of authorizing data
modifications without even realizing, early enough, the
damage they cause to the integrity of the information. A
static modeling of data, not taking into account the lifecycles
of business objects, can lead to data modifications even
though the business context does not enable it. It is therefore
necessary to take a step back from the study of the data in
order to find and counter the risks of the deterioration of
integrity in the MDM system itself. Static modeling of the
