The majority of applications need to persist (or store) data during their lifecycle. There are many ways of doing
this with an application written in Java.
You can handle the persistence (and retrieval) of data yourself using JDBC. With JDBC, you have the burden
of having to write the persistence layer yourself. This gives much control, but also creates significant work, both in
writing the code but also in testing and maintenance.
You could also use J2EE, and persist data using Entity Beans. With J2EE you hand off the persistence of
your objects to the EJB part of the J2EE server. This simplifies things, but places a major restriction on your code
in that you have to write your objects as Entity Beans. In addition J2EE adds layers that are not necessarily required
for your application.
You can use a proprietary or non-standards compliant persistence framework like TopLink, or Hibernate and
persist your plain old Java objects. You will, however, be tied in to their API and be unable to easily move your system
to an alternative implementation.
Alternatively you can use JDO, a standardised persistence interface. With JDO you can take your plain
old Java objects and just persist them as they are. It is totally transparent and very little work is required by
the developer.
To give a guide, here are a few important consideration points when choosing a persistence layer for your application.
| Feature | JDBC | Custom O/R (1) | EJB (2) | JDO |
|---|
| Standards-Driven and Portable | Yes | No | Yes | Yes |
| Choice of datastores | No | Yes | Yes | Yes |
| Simple Java objects | Yes | Yes | No | Yes |
| Out of box implementation (3) | No | Yes | No | Yes |
| Usable in stand-alone application | Yes | Yes | No | Yes |
| Usable in J2EE environment | Yes | Yes | Yes | Yes |
| Simple to unit test | Yes | Yes | No | Yes |
| Dynamic queries | Yes (4) | Yes | No | Yes |
| Primary Key generation | Yes (4) | Yes | No (5) | Yes |
| Supports inherited objects | Yes (4) | Yes | No (5) | Yes |
| Automatic Creation of schema (top-down) | No | Yes | No (5) | Yes |
| Supports existing schema (bottom-up) | Yes | Yes | Yes | Yes |
- refers to products such as Hibernate, TopLink which follow no standard.
- refers to entity beans for EJB 2.*
- refers to whether it is necessary to write the persistence yourself (e.g as with JDBC) or whether you can just
persist by simple calls.
- requires the developer to write this layer.
- some advanced EJB containers like Bea Weblogic and IBM Websphere provide support for primary key generation,
inherited objects and database schema creation as well.
