Solomon, Martin K.

In 1970, E. F. Codd published the paper "A relational model of data for large shared data banks" that introduced the relational data model for managing data. SQL, currently the most widely used relational language, was introduced in the experimental IBM database management system (DBMS) System R. Another DBMS, called INGRES, was developed at about the same time as System R (under the leadership of M. Stonebraker), and it was based on the relational database language QUEL. Since that time, inadequacies of the relational model and the need to handle more complex database issues warranted the extension of the relational model and the implementation of a DBMS that meets those needs. M. Stonebraker developed the post-relational DBMS POSTGRES based on the POSTQUEL database language. An evaluation of POSTGRES will be conducted to determine the adequacy of POSTGRES with respect to traditional relational database issues, as well as more complex issues that are considered post-relational.

This thesis describes the use and implementation of the
PEIRCE DBMS.
PEIRCE is an easy-to-use, efficient, portable and relationally
complete database management system.
The design philosophy behind PEIRCE is to provide a
powerful and efficient data manipulation language that
can be easily implemented on a variety of computers,
including microcomputers.
PEIRCE requires the user to learn only a small number of
simple, yet powerful commands. It is PEIRCE not the user
that bears the burden of efficiently implementing these
commands.
PEIRCE has been entirely written in C to enhance portability
to many computers.

In this thesis, different methods to access relational databases from an object oriented programming language like Java have been investigated. The approaches that are covered are JDBC (Java Database Connectivity), SQLJ (industry standard for embedding SQL in Java) and Java Blend (product from Sun Microsystems, that provides object-oriented way to access data in a relational database). The emphasis is given to Java Blend, as it is a new tool with a very broad scope. And arguments in support of Java Blend as a better approach compared to JDBC and SQLJ are presented.