Database research has been successful over 30 years of its history. The recent research direction in this field is two-fold: core database technology and infrastructure for information service development. While the former is well established and will continue to take the major efforts from research community and commercial vendors worldwide, the latter has been less central and thus under-developed despite its potential in the future market. With the explosion of Internet population of both individuals and development in the latter is becoming more and more significant drawing more attention than ever. Core database research and development efforts during the last two decades have been focused solely on relational data management systems. Object-oriented paradigm was eventually introduced with its powerful modeling and programming capabilities. However, due to its performance problems and implementation difficulties, it is widely observed that the mature technology and the well-established relational database market may not be displaced by object-oriented competitors (as claimed by the researches in that field). This triggered a new research and development paradigm of object-relational database management to be investigated by merging the modeling power of object-oriented system and processing efficiency of relational system. DBMS vendors and research community have announced lots of work on this issue, and they have been well accepted. In this approach, relational databases are viewed as the means of information reservoirs and underlying data management mechanism which persistently support object-oriented application. Therefore, object-relational database approach seemingly suits the purpose of storing and manipulation physical data in relational format while higher-level data modeling and application programming in an object-oriented environment. In this thesis, we present a design and implementation of an Object Database Management Group (ODMG) Object Def...