Databases Demystified


Thirty years ago, databases were found only in special research laboratories where
computer scientists struggled with ways to make them efficient and useful, and published their findings in countless research papers. Today databases are a ubiquitous
part of the information technology (IT) industry and business in general. We directly
and indirectly use databases every day—banking transactions, travel reservations,
employment relationships, web site searches, purchases, and most other transactions are recorded in and served by databases.
As with many fast-growing technologies, industry standards have lagged behind
the development of database technology, resulting in a myriad of commercial products, each following a particular software vendor’s vision. Moreover, a number of
different database models have emerged, with the relational model being the most
prevalent. Databases Demystified examines all of the major database models, including hierarchical, network, relational, object-oriented, and object-relational.
However, Databases Demystified concentrates heavily upon the relational and object-relational models because these are the mainstream of the IT industry and will
likely remain so in the foreseeable future.
The most significant challenge in implementing a database is designing the structure of the database correctly. Without a thorough understanding of the problem the
database is intended to solve, and without knowledge of the best practices for organizing the required data, the implemented database becomes an unwieldy beast that
requires constant attention. Databases Demystified focuses on transformation of requirements into a working database model with special emphasis on a process called
normalization, which has proven to be an effective technique for designing relational databases. In fact, normalization can be applied successfully to other database