Published on ONJava.com (http://www.onjava.com/)
 See this if you're having trouble printing code examples

O'Reilly Book Excerpts: Java Data Objects

JDO Architectures

Related Reading

Java Data Objects
By David Jordan, Craig Russell

by David Jordan and Craig Russell

Editor's Note: In this excerpt from Java Data Objects, authors David Jordan and Craig Russell provide a high-level overview of the architectural aspects of JDO, as well as examples of environments in which JDO can be used.

One of JDO's primary objectives is to provide you with a transparent, Java-centric view of persistent information stored in a wide variety of datastores. You can use the Java programming model to represent the data in your application domain and transparently retrieve and store this data from various systems, without needing to learn a new data-access language for each type of datastore. The JDO implementation provides the necessary mapping from your Java objects to the special datatypes and relationships of the underlying datastore. Chapter 4 discusses Java modeling capabilities you can use in your applications. This chapter provides a high-level overview of the architectural aspects of JDO, as well as examples of environments in which JDO can be used. We cannot enumerate all such environments in this book, because JDO is capable of running in a wide variety of architectures.

A JDO implementation is a collection of classes that implement the interfaces defined in the JDO specification. The implementation may be provided by an Enterprise Information System (EIS) vendor or a third-party vendor; in this context, we refer to both as JDO vendors. A JDO implementation provided by an EIS vendor will most likely be optimized for the specific EIS.

The JDO architecture simplifies the development of scalable, secure, and transactional JDO implementations that support the JDO interface. You can access a wide variety of storage solutions that have radically different architectures and data models, but you can use a single, consistent, Java-centric view of the information from all the datastores.

The JDO architecture can be used to access and manage data contained in local storage systems and heterogeneous EISs, such as enterprise resource planning (ERP) systems, mainframe transaction processing systems, and database systems. JDO was designed to be suitable for a wide range of uses, from embedded small-footprint systems to large-scale enterprise application servers. A JDO implementation may provide an object-relational mapping tool that supports a broad array of relational databases. JDO vendors can build implementations directly on the filesystem or as a layer on top of a protocol stack with multiple components.

JDO has been designed to work in three primary environments:

You can focus on developing your application's business and presentation logic without having to get involved in the issues related to connecting to a specific EIS. The JDO implementation hides the EIS-specific issues, such as datatype mapping, relationship mapping, and the retrieval and storage of data. Your application sees only a Java view of the data, organized as classes using native Java constructs. EIS-specific issues are important only during deployment of your application.

In a nonmanaged environment, you do not rely on the managed services of security, transaction, and connection management offered by a middle-tier application server. Chapters through cover the uses of JDO in a nonmanaged environment, most of which also apply to a managed environment.

When JDO is deployed in a managed environment, it uses the J2EE Java Connector Architecture, which defines a set of portable, scalable, secure, and transactional mechanisms for integrating an EIS with an application server. These mechanisms focus on important aspects of integration with heterogeneous systems: instance management, connection management, and transaction management. The Java Connector Architecture enables a standard JDO implementation to be pluggable across application servers from multiple vendors.

Managed environments also provide transparency for application components' use of system-level mechanisms--distributed transactions, security, and connection management--by hiding the contracts between JDO implementation and the application server. Chapter 16 covers the use of JDO in the web server environment. Chapter 17 explains how to use JDO to provide persistence services in a J2EE application-server environment, which supports the Enterprise JavaBeans (EJB) architecture.

Multiple JDO implementations--possibly multiple implementations per type of EIS or local storage--can be plugged into an application server concurrently, or they can be used directly in a two-tier or embedded architecture. JDO also allows a persistent class to be used concurrently with multiple JDO implementations in the same Java Virtual Machine (JVM) or application-server environment. This enables application components--deployed on a middle-tier application server or client-tier--to access the underlying datastores using the same consistent, Java-centric view of data.

The persistent classes that you define can migrate easily from one environment to another. This also allows you to debug persistent classes and parts of your application code in a simple one- or two-tier environment and deploy them in another tier of the system architecture.

Architecture Within Application JVM

JDO supports a variety of architectures within the application's JVM context. Your application can have one or multiple PersistenceManagers accessing the same or different datastores concurrently. Each PersistenceManager has its own persistent instance cache and its own associated Transaction instance, which manages a distinct transactional context. A JDO implementation may also maintain a shared cache of instances (not visible to applications) to optimize the application's access of data in the datastore.

Single PersistenceManager

The simplest JDO application architecture has a single PersistenceManager, as illustrated in Figure 3-1. A PersistenceManager is the primary interface used by the application to access persistent services. It is an interface that is implemented by an instance of the JDO implementation. The persistent instances are managed in a cache, where they are used directly by the application. The JDO implementation manages the persistent instances both by using application control (e.g., using PersistenceManager and Query methods), and transparently (when the application accesses a field that is not loaded). The cache contains other artifacts, used to track the identity and state of the instances, but these artifacts are not visible to the application. Whenever we mention the cache, we are referring to the cache of persistent instances.

Figure 3-1. Application using a single PersistenceManager to access a datastore

The application cache is not a specific region of memory, as Figure 3-1 might imply; it is simply part of the JVM's object heap. Each persistent class has a field, named jdoStateManager, added by the enhancer to reference a StateManager. The StateManager manages the field values and lifecycle state of the instance, and has a reference to its associated PersistenceManager. A PersistenceManager may use one or more StateManagers; this detail is implementation-specific. The jdoStateManager field for any instance being managed (either a persistent or transient transactional instance) is set to reference a StateManager; otherwise, the jdoStateManager field is null.

A persistent instance in the cache can directly reference other persistent instances in the same cache. You can navigate from one instance to another using standard Java syntax. Instances of transient classes (for example, your application class) can also reference these persistent instances. A persistent instance in the cache can also reference transient instances of both persistent and transient classes. The persistent classes themselves are responsible for managing references to transient instances; the JDO implementation does not manage these references.

Figure 3-2 shows the relationships between the persistent instances, the StateManager, and the PersistenceManager. Each persistent instance contains a reference to a StateManager, which can manage one or more persistent instances. Each StateManager contains a reference to its PersistenceManager, which can manage one or more StateManagers. Each PersistenceManager contains a reference to its PersistenceManagerFactory, which can manage one or more PersistenceManagers. Each PersistenceManager can manage one transaction serially, and contains a reference to its Transaction instance. The PersistenceManager uses a StoreManager to interact with the datastore; this relationship is not defined by the JDO specification.

Figure 3-2. UML diagram of persistent instance cache

Multiple PersistenceManagers Accessing the Same Datastore

You can instantiate multiple PersistenceManagers in your application from the same or different PersistenceManagerFactorys. Figure 3-3 illustrates an application with two PersistenceManagers from the same PersistenceManagerFactory.

Figure 3-3. Application with multiple PersistenceManagers

Each PersistenceManager manages its own transaction context and application cache. In this particular example, both PersistenceManagers access the same datastore and are from the same JDO implementation. This is the typical architecture for managed environments where different instances of the same component access the same datastore via different PersistenceManagers.

Both PersistenceManagers may have the same datastore instance in their caches, represented by different persistent instances. This architecture provides for transactional isolation of changes made to the same datastore instance by different transactions.

Multiple PersistenceManagers Accessing Different Datastores

Figure 3-4 illustrates PersistenceManagers accessing different datastores. These PersistenceManagers could be from the same or different implementations. For example, one datastore may be a relational database and the other an object database. Due to JDO's binary-compatibility contract (covered in Chapter 6), PersistenceManagers from different implementations can manage different instances of the same persistent classes. JDO is the first database-interface technology to offer this high level of portability across database architectures.

Figure 3-4. Application with multiple JDO implementations

Shared Implementation Cache

In addition to the application cache, some JDO implementations also maintain their own persistent instance cache that sits between the application cache and the datastore. Your application does not have access to this implementation cache. Its role is to cache the state of objects from the datastore in memory, so they can be provided to the application without requiring access to the datastore. Use of caches can result in significant performance improvements. A shared implementation cache is most useful when you use nontransactional access, covered in Chapter 14, or optimistic transactions, covered in Chapter 15. When you use datastore transactions, the shared cache is usually bypassed.

Shared implementation cache within a single JVM

Figure 3-5 illustrates a shared implementation cache that is managed within a single JVM. It allows each of the PersistenceManagers to quickly access the state of objects that have been accessed from the same datastore.

Figure 3-5. Implementation of a shared cache for transactions accessing the same datastore

For example, if one PersistenceManager accesses a particular instance, the implementation needs to read the instance from the datastore. But if the other PersistenceManager then accesses the same instance, the implementation can use the data in the shared implementation cache and avoid having to access the datastore.

Shared implementation cache distributed among JVMs

Several JDO implementations provide a distributed cache architecture, which allows them to migrate the state of objects between JVMs. Figure 3-6 illustrates this architecture.

Figure 3-6. Implementation use of distributed, synchronized caches

Again, the goal with these implementations is to avoid a datastore access whenever possible. For some systems where multiple applications may access the same objects, these implementations demonstrate substantial performance improvements.

Datastore Access

We have explored the architecture in the application's JVM and discussed the application cache and implementation cache. Now let's examine the architectures of JDO implementations. We'll discuss each type of datastore separately.

These architectures don't affect your application's programming model, but they affect the configuration of the environment in which your application executes. In particular, the ConnectionURL property of the Properties instance used to construct the PersistenceManagerFactory refers to a local or remote datastore.

Direct Access of Filesystem or Local Datastore

Some JDO implementations store the objects directly in a local filesystem or datastore. Figure 3-1 illustrates this architecture. There is only a single process context in this architecture. The JDO implementation uses the Java I/O classes directly to manage the storage of the objects in a file. The JDO Reference Implementation implements this architecture, as do some object databases.

Remote Access of a JDO Server

Some JDO implementations connect to a separate server that manages the datastore, as illustrated in Figure 3-7. The JDO Reference Implementation implements this architecture, as do most object databases. In this particular example, the JDO implementation itself provides a server built specifically for object storage, which then manages the filesystem directly. The component that executes in the same JVM as the JDO implementation and communicates with the remote server is called a resource adapter. The protocols between the client JVM and the JDO Server are vendor-specific.

Figure 3-7. Client access of a JDO server

Remote Access of a SQL Datastore

Figure 3-8 illustrates the use of a relational database server for object storage. This is the most common architecture used by current commercial JDO implementations. Since the application is written in Java, the JDO implementation uses JDBC to communicate with the database server. When you deploy your application, you use a proprietary tool supplied by the JDO vendor to map your application's Java objects to tables in the relational database. Some JDO implementations use your application's persistent object model to create the relational schema for you.

Figure 3-8. Client access of a SQL datastore

The relational vendor or a third party provides a JDBC driver to communicate with the database, using protocols specific to the database. The JDBC driver is the resource adapter in this architecture.

Since the JDBC interface is well defined, this architecture offers a high degree of portability. JDO implementations have been written to use a variety of datastores that provide a JDBC driver implementation. While the JDBC interface is standard, the SQL data manipulation language, used by the relational databases, varies considerably; the JDO implementation hides these differences from JDO applications.

System Architectures with a JDO Application

Now we'll examine where JDO objects and application logic can be placed relative to an application's overall system architecture, including both managed and nonmanaged environments. In the remaining examples in this chapter, we don't show the details of how the JDO implementation manages the storage for the persistent instances.

JDO Rich Client with Local Datastore

The simplest form of system architecture is a one- or two-tier application that may be executed from the command line, from a shell script, or via a graphical user interface. We refer to the application as a rich client to distinguish it from a browser that simply displays HTML and executes applets. The application uses local filesystem and JDO persistent services directly.

JDO Applications in a Web Server

Figure 3-9 illustrates how an application can use JDO to provide persistent services to the implementation of a web servlet or JavaServer Pages (JSP). When using JSP pages, the application typically will use JDO in one of two ways: by calling JDO's APIs directly in Java, or using a JSP tag library to abstract the JDO API (similar to the way the JSP Standard Tag Library abstracts the JDBC API).

Figure 3-9. JDO application running in a web server

With this architecture, the servlet/JSP page gets data from the browser in the form of strings from an HTTP doGet( ) or doPost( ) request and uses JDO to implement the request. Your application may use the Struts framework to implement the servlets and JSP pages in this architecture. We will discuss the web-server access patterns in detail in Chapter 16.

JDO Applications as Web Services

Figure 3-9 also illustrates the use of JDO as the persistence implementation for a web server implementation of a web services endpoint. The web server may register the service using UDDI and a registry service, and clients may find the service via the same registry.

A web server implementation uses a servlet to implement the service endpoint. The servlet can use the JDO API for the persistent service, exactly as it does for servicing HTTP requests. The primary difference between SOAP and standard HTTP is that with SOAP requests, the message data in the HTTP message is formatted as SOAP XML instead of get/post data.

Rich Client Connecting to Application Server with EJB Components

Figure 3-10 illustrates a rich client connecting directly to an application server using EJB beans. This architecture typically is implemented behind the firewall of a company, as it directly exposes enterprise services to clients. The clients use the JNDI services of the J2EE client container to look up services by name (including EJB beans) and to connect to the server via RMI/IIOP or a proprietary protocol. Alternatively, a client may use SOAP protocols to access the middle-tier server.

Figure 3-10. Rich-client connection to an application server using EJB beans

The EJB components inside the EJB container use other EJB components to implement their services. They use a combination of JDBC and JDO to access persistent services. Session beans and message-driven beans use JDO and JDBC directly. Entity beans use JDO transparently (the container implements CMP entity beans using JDO but does not expose JDO as an API to the CMP developer).

Web Server with EJB Server

Figure 3-11 illustrates servlets and JSP pages that use the services of an EJB container to implement the business logic of an enterprise application. The EJB beans executing inside the EJB container use JDO as their persistence service. The web and EJB containers often reside in the same JVM in this architecture, even though they represent different tiers of the architecture.

Figure 3-11. Servlets and JSP pages access services of the EJB container

EJB Session Beans Using Session Bean Fašades

Figure 3-12 illustrates the session bean delegating parts of the business logic to session bean fašades that use JDO as their implementation. This architecture allows location transparency among the components. For example, if the session bean that interacts directly with clients delegates part of the functionality to other session-bean components, this architecture allows the other components to be located in different machines. Chapter 17 describes this architecture in detail.

Figure 3-12. EJB session beans using session bean delegates

JDO Providing Container-Managed Persistence

As a side note, an EJB server may implement J2EE container-managed persistence (CMP) entity beans using JDO as the persistence layer. The J2EE components and the users of these components are unaware that JDO is used for the implementation of the persistence service.

David Jordan founded Object Identity, Inc. to provide Java Data Objects (JDO) consulting and training services. David is also a coauthor of O'Reilly's book on Java Data Objects, with Craig Russell.

Craig Russell is the specification lead for JDO at Sun Microsystems.

Return to ONJava.com.

Copyright © 2009 O'Reilly Media, Inc.