Information Technology Reference
In-Depth Information
Because today's businesses have so many elements, they must keep data organized so that it
can be used effectively. A database should be designed to store all data relevant to the business
and provide quick access and easy modification. Moreover, it must reflect the business pro-
cesses of the organization. When building a database, an organization must carefully consider
these questions:
•
Content.
What data should be collected and at what cost?
•
Access.
What data should be provided to which users and when?
•
Logical structure.
How should data be arranged so that it makes sense to a given user?
•
Physical organization.
Where should data be physically located?
Data Modeling
Key considerations in organizing data in a database include determining what data to collect
in the database, who will have access to it, and how they might want to use the data. After
determining these details, an organization can create a database. Building a database requires
two different types of designs: a logical design and a physical design. The
logical design
of a
database is an abstract model of how the data should be structured and arranged to meet an
organization's information needs. The logical design involves identifying relationships
among the data items and grouping them in an orderly fashion. Because databases provide
both input and output for information systems throughout a business, users from all func-
tional areas should assist in creating the logical design to ensure that their needs are identified
and addressed.
Physical design
starts from the logical database design and fine-tunes it for
performance and cost considerations (such as improved response time, reduced storage space,
and lower operating cost). For example, the database administrator at Intermountain Health-
care in Salt Lake City, Utah, combined the databases of 21 hospitals and 100 clinics into one
integrated system, saving the organization the cost of dozens of servers, and providing new
knowledge of the DBMS. For example, the logical database design might need to be altered
so that certain data entities are combined, summary totals are carried in the data records
rather than calculated from elemental data, and some data attributes are repeated in more
than one data entity. These are examples of
planned data redundancy
, which improves the
system performance so that user reports or queries can be created more quickly.
One of the tools database designers use to show the logical relationships among data is a
data model. A
data model
is a diagram of entities and their relationships. Data modeling
usually involves understanding a specific business problem and analyzing the data and in-
formation needed to deliver a solution. When done at the level of the entire organization,
this is called enterprise data modeling.
Enterprise data modeling
is an approach that starts
by investigating the general data and information needs of the organization at the strategic
level, and then examines more specific data and information needs for the various functional
areas and departments within the organization. Various models have been developed to help
managers and database designers analyze data and information needs. An entity-relationship
diagram is an example of such a data model.
Entity-relationship (ER) diagrams
use basic graphical symbols to show the organization
of and relationships between data. In most cases, boxes in ER diagrams indicate data items
or entities contained in data tables, and diamonds show relationships between data items and
entities. In other words, ER diagrams show data items in tables (entities) and the ways they
are related.
ER diagrams help ensure that the relationships among the data entities in a database are
correctly structured so that any application programs developed are consistent with business
operations and user needs. In addition, ER diagrams can serve as reference documents after
a database is in use. If changes are made to the database, ER diagrams help design them.
Figure 5.4 shows an ER diagram for an order database. In this database design, one salesperson
serves many customers. This is an example of a one-to-many relationship, as indicated by
planned data redundancy
A way of organizing data in which the
logical database design is altered so
that certain data entities are com-
bined, summary totals are carried in
the data records rather than calcu-
lated from elemental data, and
some data attributes are repeated in
more than one data entity to improve
database performance.
data model
A diagram of data entities and their
relationships.
enterprise data modeling
Data modeling done at the level of
the entire enterprise.
entity-relationship (ER)
diagrams
Data models that use basic graphi-
cal symbols to show the organiza-
tion of and relationships between
data.
