In the previous chapter, we discussed ways of using JDBC to access a database. In the context of object-oriented software design, it is often required to not only store simple values in a database but to also map the objects and their relations to the tables.
This chapter is an introduction to the concept of Object-relational mapping (ORM) and the Hibernate framework as an implementation of ORM. The first part describes the overall architecture of Hibernate, and gives guidelines for installing and configuring the framework.
Later in this chapter, the basic usage of Hibernate, such as performing CRUD operations, fetching strategies, and association mapping, will be covered.
Finally, the end of the chapter introduces advanced techniques such as caching, pooling, and using table partitioning in conjunction with Hibernate.
While relational databases represent data in a tabular form, object-oriented programming languages organize the data in a graph of objects that reference each other. This results in different approaches for accessing and manipulating stored data.
An ORM framework translates between tabular and object representation, and creates a kind of virtual object database that can be used from within an application.
Hibernate is an open-source persistence framework for Java. Beside the ORM module, which will be discussed in the context of this chapter, it provides additional modules and tools for accessing the NoSQL data-stores, adding full-text search, or for validating data.
The Hibernate ORM module takes care of the mapping between the database tables and Java classes, and largely, reduces the amount of code needed to implement data persistence.
Hibernate inserts a layer between the Java application and JDBC, as shown in the following diagram. Along with JDBC, it uses the Java Naming and Directory Interface (JNDI) and the Java Transaction API (JTA) to allow integration with the J2EE application servers:
Hibernate ORM architecture