Chapter 11. Connection Pools and Data Sources

Table of Contents

Overview
Application Servers: ConnectionPoolDataSource
Applications: DataSource
Tomcat setup
Data Sources and JNDI

JDBC 2 introduced standard connection pooling features in an add-on API known as the JDBC 2.0 Optional Package (also known as the JDBC 2.0 Standard Extension). These features have since been included in the core JDBC 3 API. The PostgreSQLJDBC drivers support these features if it has been compiled with JDK 1.3.x in combination with the JDBC 2.0 Optional Package (JDBC 2), or with JDK 1.4 or higher (JDBC 3). Most application servers include the JDBC 2.0 Optional Package, but it is also available separately from the Sun JDBC download site.

The JDBC API provides a client and a server interface for connection pooling. The client interface is javax.sql.DataSource, which is what application code will typically use to acquire a pooled database connection. The server interface is javax.sql.ConnectionPoolDataSource, which is how most application servers will interface with the PostgreSQLJDBC driver.

In an application server environment, the application server configuration will typically refer to the PostgreSQLConnectionPoolDataSource implementation, while the application component code will typically acquire a DataSource implementation provided by the application server (not by PostgreSQL™).

For an environment without an application server, PostgreSQL™ provides two implementations of DataSource which an application can use directly. One implementation performs connection pooling, while the other simply provides access to database connections through the DataSource interface without any pooling. Again, these implementations should not be used in an application server environment unless the application server does not support the ConnectionPoolDataSource interface.