Lecture 22. Object-oriented DBMS

The direction of object-oriented databases (OODB) arose relatively long. Publications appeared in the mid-1980s. However, this area has been developing most actively in recent years. The number of publications and implemented commercial and experimental systems is increasing every year.

The emergence of the OODB direction is primarily determined by practical needs: the need to develop complex information application systems for which the technology of the previous database systems was not entirely satisfactory.

Of course, OOBD did not arise from scratch. The corresponding basis is provided by both previous work in the field of databases, and the long-developing areas of programming languages ​​with abstract data types and object-oriented programming languages.

As for the connection with previous works in the field of databases, in our opinion, the most dramatic impact on the work in the field of RDBs is the development of relational DBMS and the following database chronologically behind them, which support the management of complex objects. In addition, an exceptional influence on the ideas and concepts of OODB and, as it seems, the whole object-oriented approach had an approach to semantic data modeling. Developing in parallel with OODD directions of deductive and active databases also have a sufficient effect.

Among the languages ​​and programming systems, Smalltalk had the greatest primary influence on OODB. This language itself is not fully pioneering, although it introduced new terminology, which is now the most common in object-oriented programming. In fact, Smalltalk is based on a number of previously put forward concepts.

A large number of published works does not mean that all OODB problems are completely solved. As noted in the Manifesto by a group of leading scientists involved in OODB, the current situation with OODB resembles the situation with relational systems of the mid-1970s. If there are a large number of experimental projects (and even commercial systems), there is no generally accepted object-oriented data model, and not because there is not a single developed full model, but because there is no general agreement on the adoption of any model. In fact, there are more specific problems related to the development of declarative query languages, the execution and optimization of queries, the formulation and maintenance of integrity constraints, access synchronization and transaction management, etc.

The theme of OOBD is very broad, the volume of this lecture does not allow to consider all the questions. Nevertheless, we will try in a systematic manner to analyze the most important aspects of OODD.