Databases Reference
In-Depth Information
administration function to set standards regarding data and its use. One example
of standards is controlling the way that attribute names, table names, and other
data-related names are formed. Attribute names must be meaningful and consistent.
The company can't have its human resources department use Serial Number as
the attribute name for employee numbers while at the same time its manufacturing
department uses it for finished product serial numbers. Similarly, there is a problem
if the human resources department tries to use Serial Number and Employee Number
in different tables to represent the employee number. Another example of standards
setting is insisting on consistency in the way the programs that access the database
are written, especially in regard to the database call instructions. Care here can
help to prevent database-call-related performance problems, as well as to ease
maintenance by having standard, readily understood instructions.
Data standards
also come into play in the IS interactions between companies
in supply chains. When data is exchanged using electronic data interchange (EDI)
technology, adjustments have to be made to take into account attribute structures
and other differences in the information systems of the two companies involved.
Liaison to Systems Analysts and Programmers
In the role of liaison to application developers, data administrators (often called ''data
analysts'' in this role) are responsible for providing support to the systems analysts
and programmers in all matters concerning the data needed by an application.
During the systems analysis phase of application development, the support may
include help in determining what data is needed for the application and which of
the data items needed for the application already exist in the active database.
Another aspect of such liaison activity, which is really a topic in itself, is
the question of database design. Data analysts are generally involved in database
design at some level, but deciding exactly what that level of involvement should
be depends on a number of factors. In an IS environment in which the data
administration organization is very strong and in which there is a significant amount
of data sharing among different applications and different functional areas of
the company, the data analysts may do all of the logical database design work
themselves. Here again, they can stand as an impartial group creating the best
design for the overall good of all of the users. The other choice is for the application
developers to do the database design with either active consultation by the data
analysts, or approval responsibility after the fact by the data analysts. In the active
consultation role, the data analysts lend their expertise to the effort, as well as
determining how the new data should mesh with data in the existing database,
if there is to be such a merging. In the approval role, the application developers
(usually the lead programmers for this activity) design the database, which is then
shown to the data analysts for discussion and approval.
Training
In some companies, data administration is responsible for training all those in the
company who need to understand the company's data and, in some cases, the DBMS
environment. Management personnel should understand why the database approach
is good for the company and for their specific individual functions. Users must
understand why the shared data is secure and private. Application developers must
be given substantial training in how to work in the database environment, including
Search WWH ::
Custom Search