Information Technology Reference
In-Depth Information
It is quite possible that the overall data content has to be loaded from different
source systems. In this case the exact sequence has to be established since interde-
pendencies between individual uploads may exist.
7.2.4 Consulting Business Units
Technical and business relevant customising can be done in parallel. Sometimes
technical assumptions have to be done when setting up the basic configuration.
These may include for example table configurations, data buffers, configuration
parameters for job streams, logical addressing, data paths etc.
Business relevant customising is very specific to the application and on top of
this depends on business processes regarding the company
s business and its
internal organisation. One important component concerns the general rules for
access rights in view of the technical possibilities at hand. This comprises among
others: who has the right to read which data, change them, to execute which
function etc.? This not only refers to the access rights as such but in consequence
also to menu and transaction control: the end user should be offered only such
functions which he needs for his job.
Other special aspects regard locking mechanisms on data items, descriptive
transaction codes, number ranges, accounting keys, descriptive texts, value ranges,
client structures, non standard reports and others. Since this may comprise a
multitude of parameters depending on the application in question it is not possible
to present a comprehensive list here.
In large corporations the business customising normally cannot be set up by one
single person but has to be worked out in consent from the inputs from different
business units. The effort in the preparation phase should not be underestimated.
'
7.2.4.1
Interfaces
Special attention regarding customising requirements is needed with respect to
interfaces. Figure
7.6
shows this relationship.
Only very rarely do the data formats on both sides of an interface correspond as
1:1 for linked up systems. Often interface data records on the exporting system are
composed from different table entries providing the information content. These
data are then presented in an export format at the interface itself.
When importing, the corresponding system takes the information it needs and
converts it in such a way as to match its own data model. Therefore a correspon-
dence between both systems has to be created. This happens via so called mapping
tables.
Mapping tables are filled with appropriate format defaults or required contents
respectively for each data attribute to be transferred. In this way all export data
records are generated. On the opposite side the import mapping table converts the
