Study On The Java Persistence Layer Design

With the rapid development of computer technology and network technique, the data sharing has been the demand of the Information Age. In order to implement the data sharing, data should not only be accessible, but also be persistent. When appling object-oriented programming method for processing and showing data, data and its processing has been wrapped into a lot of object ,so the study of data persistence turned to the object's. There are many ways to persist data in recent years, such as file system, object oriented database system (OODBS), relation database management system (RDBMS), etc. Because the RDBMS is the most mature data persistent mechanism, it has been used in a plumping majority of Management Information Systems. Generally speaking, applications using the RDBMS as the data persistence mechanism manipulate the RDBMS by JDBC on the logic layer, it makes the manipulation of the object and RDBMS mixed together, lower the maintainability of the system.By analyzing five data persistence methods at home and abroad, such as JDBC, ORM, a coarse-grained component based JDBC has been designed, it not only provides interfaces for RDBMS's common operation, but features such as transaction management and calling to stored procedures, which are independent with business logic. The component can be used separately to do the data persistence layer design, or with the J2EE core modes, for example DAO, to make the pensistence object oriented. It has been applied to design the data persistence layer in project of"Undergraduate Students'Thesises Management System"to prove the usability, concision and reusability of the component. Then improved the persistence layer by appling DAO to show its flexibility and extensibility. Finally, compared its capabilities with Hibernate under simular conditions of hardware and software, proving that the component is with higher valued use.