Allocation of Classes in Distributed Object-Oriented Databases

In distributed object-oriented databases(DOODB), objects are distributed in different sites on a communication network. A class fragmentation that divides a class into several fragments is needed for improving performance and for reducing repetition and duplication of data transmission. It is proposed the vertical class fragmentation to reflect the characteristics of object-oriented database such as method, inheritance and composite-object. In this paper, we define the objective function for allocation considering system to save cost including storage, query processing and communication and implemented it using genetic algorithm.