Java Persistence/What is JPA?
The Java Persistence Architecture API (JPA) is a Java specification for accessing, persisting, and managing data between Java objects / classes and a relational database. JPA was defined as part of the EJB 3.0 specification as a replacement for the EJB 2 CMP Entity Beans specification. JPA is now considered the standard industry approach for Object to Relational Mapping (ORM) in the Java Industry.
JPA itself is just a specification, not a product; it cannot perform persistence or anything else by itself. JPA is just a set of interfaces, and requires an implementation. There are open-source and commercial JPA implementations to choose from and any Java EE 5 application server should provide support for its use. JPA also requires a database to persist to.
JPA allows POJO (Plain Old Java Objects) to be easily persisted without requiring the classes to implement any interfaces or methods as the EJB 2 CMP specification required. JPA allows an object's object-relational mappings to be defined through standard annotations or XML defining how the Java class maps to a relational database table. JPA also defines a runtime EntityManager API for processing queries and transaction on the objects against the database. JPA defines an object-level query language, JPQL, to allow querying of the objects from the database.
JPA is the latest of several Java persistence specifications. The first was the OMG persistence service Java binding, which was never very successful; I'm not sure of any commercial products supporting it. Next came EJB 1.0 CMP Entity Beans, which was very successful in being adopted by the big Java EE providers (BEA, IBM), but there was a backlash against the spec by some users who thought the spec requirements on the Entity Beans overly complex and overhead and performance poor. EJB 2.0 CMP tried to reduce some of the complexity of Entity Beans through introducing local interfaces, but the majority of the complexity remained. EJB 2.0 also lacked portability, in that the deployment descriptors defining the object-relational mapping were not specified and were all proprietary.
This backlash in part led to the creation of another Java persistence specification, JDO (Java Data Objects). JDO obtained somewhat of a "cult" following of several independent vendors such as Kodo JDO, and several open-source implementations, but never had much success with the big Java EE vendors.
Despite the two competing Java persistence standards of EJB CMP and JDO, the majority of users continued to prefer proprietary api solutions, mainly TopLink (which had been around for some time and had its own POJO API) and Hibernate (which was a relatively new open-source product that also had its own POJO API and was quickly becoming the open-source industry standard). The TopLink product formerly owned by WebGain was also acquired by Oracle, increasing its influence on the Java EE community.
The EJB CMP backlash was only part of a backlash against all of Java EE which was seen as too complex in general and prompted such products as the Spring container. This led the EJB 3.0 specification to have a main goal of reducing the complexity, which led the spec committee down the path of JPA. JPA was meant to unify the EJB 2 CMP, JDO, Hibernate, and TopLink APIs and products, and seems to have been very successful in doing so.
Currently most of the persistence vendors have released implementations of JPA confirming its adoption by the industry and users. These include Hibernate (acquired by JBoss, acquired by Red Hat), TopLink (acquired by Oracle), and Kodo JDO (acquired by BEA, acquired by Oracle). Other products that have added support for JPA include Cocobase (owned by Thought Inc.) and JPOX.