When you create a PersistenceManagerFactory / EntityManagerFactory you define the connection URL, driver name, and the username/password to use. This works perfectly well but does not "pool" the connections so that they are efficiently opened/closed when needed to utilise datastore resources in an optimum way. DataNucleus allows you to utilise a connection pool using BoneCP to efficiently manage the connections to the datastore.

BoneCP is a third-party library providing connection pooling. This is accessed by specifying the persistence property datanucleus.connectionPoolingType . To utilise BoneCP-based connection pooling we do this

// Specify our persistence properties used for creating our PMF/EMF
Properties props = new Properties();
properties.setProperty("javax.jdo.PersistenceManagerFactoryClass",
                "org.datanucleus.jdo.JDOPersistenceManagerFactory");
properties.setProperty("datanucleus.ConnectionDriverName","com.mysql.jdbc.Driver");
properties.setProperty("datanucleus.ConnectionURL","jdbc:mysql://localhost/myDB");
properties.setProperty("datanucleus.ConnectionUserName","login");
properties.setProperty("datanucleus.ConnectionPassword","password");
properties.setProperty("datanucleus.connectionPoolingType", "BoneCP");

So the PMF / EMF will use connection pooling using BoneCP. To do this you will need the BoneCP JAR (and SLF4J, google-collections) to be in the CLASSPATH.

You can also specify persistence properties to control the actual pooling. The currently supported properties for BoneCP are shown below

# Pooling of Connections
datanucleus.connectionPool.maxPoolSize=5
datanucleus.connectionPool.minPoolSize=3

# Pooling of PreparedStatements
datanucleus.connectionPool.maxStatements=20