BoneCP is a Java JDBC connection pool implementation that is tuned for high performance by minimizing lock contention to give greater throughput for your applications. It currently beats all known connection pools, including C3P0 and DBCP, as shown on the benchmark section on the site.
|Tags||jdbc connection pool hibernate spring MySQL Oracle MSSQL postgres Database c3p0 dbcp primrose benchmarks|
|Operating Systems||OS Independent|
Release Notes: Plenty of new features and bug fixes were added
Release Notes: An architectural change was done to make the pool faster. Bugs were fixed. A bunch of new features were added.
Release Notes: This is mostly a maintenance release. Fixes were made for idleMaxAge/idleConnectionTester, multiple nodes in XML configuration, and an NPE on getConnection timeout. Batch statements are logged properly. The onAcquireFail hook was enhanced, and a tiny bug was fixed in the process. A driver properties option was added. The connection is no longer closed off when shutting down. A bug in the close connection watch debug tool was fixed. All release helpers are terminated when shutting down.
Release Notes: A driver properties option was added. XML-based configuration support was added. onAcquireFail was enhanced. A connection timeout configuration option was added. Full JDK5 compatibility was added. Internal statements are exposed. The onConnectionException hook was added. Disabling connection tracking is now allowed. Batch statements are now logged properly. A lazy init race was fixed. A potential race is now avoided. Statics were removed from finalizers. A potential bug in adding a new connection was fixed. There are now 125 unit tests, which provide 100% unit test coverage.
Release Notes: This release adds an onConnectionException hook to allow custom action when a connection fails. It exposes internal statements, allows for getConnection timeouts, and allows code to be built with JDK5. It removes the log4j dependency from pom, silences a warning, and makes config.toString() report timeouts in minutes. A potential race condition in PoolWatch is avoided. Static has been removed from finalizer to allow for multiple create/destroy pool cycles. A startup hang on single processor where lazyInit = true has been fixed. There are 121 unit tests and 100% test coverage.