Geography Reference
In-Depth Information
Security number, employee number, or student number. In the case of automobiles, the key code might
be the vehicle identification number (VIN).
To illustrate why multiple tables might be used, suppose that, in our relational database, it is possible for
one person to own more than one automobile. If the entire database is all in one table, then the names and
addresses of each multicar owner must be repeated, which increases the amount of storage required. If a
multicar owner gets a new street address, several records in the database table would have to be changed.
Further, some of the data about owners might be located in another database and might be private.
The elegant answer to these problems is to have several tables that contain the information, divided to
minimize the repetitions. In the automobile example, it might be that only the owner's identification
number is stored in the record of the car itself. This number could be the key field in the database table
containing information about owners. If some of the owner information is private, it could be stored in a
separate table as well, with the same key field. The database could be set up so that this table could not
be accessed by the automobile table.
Relational database tables that meet certain requirements of efficiency are said to be in first normal form
(1NF), second normal form (2NF), and so on. The higher the number, the more efficient the database.
In summary, it is useful to partition the information in a relational database into a number of tables. Such
partitioning can
Reduce the amount of redundant data stored
❏
Aid in updating the database
❏
Reduce the chances of inconsistency and instability in the database
❏
Aid in protecting private or sensitive data
❏
Languages have been developed for retrieving information from a relational database. One is the
Structured Query Language (SQL), developed by the American National Standards Institute (ANSI). You
use it to execute queries of a database. Vendors of relational database management systems also have
developed proprietary languages.
24
Esri products interface with a number of RDBMS from
various sources.
Databases—What's Meant by “Relational”
What you have seen so far is one aspect of the RDB: a two-dimensional table in which you store entities
as rows and attributes as columns. What is it that makes the relational database such a powerful
approach to storing information? One answer lies in the fact that a RDB can be much more than a single
table. Usually it is a number of tables that are related to one another, as previously discussed, that
provide for efficiency, flexibility, and ease of updating.
Table 1-1 illustrates a database that might be created by a motor vehicles licensing department. Parts of
the database connect to other databases formed by other government departments. Shown in Table 1-1
are nine relational database tables; the RDB table name is on the top line; the attribute names are on the
second line. None of the tens of thousands of cells is shown.
24
Some names of past and present relational database management system software products are Access, dBASE,
Informix, Ingres, Oracle, SQL Server, and Sybase.
Search WWH ::
Custom Search