Apache BookKeeper was built on top of Apache ZooKeeper, a centralized distribution server where information and configurations about distributed applications can be managed.
BookKeeper uses ZooKeeper's underlying code to manage, centralize and distribute log messages inside a network of computers.
The project consists of two smaller parts. One is the distributed logging service called BookKeeper and the second is a publish/subscribe server dubbed Hedwig.
These two are used together to track changes made to a state machine and then replicate them across a network of distributed computers.
If the state machine (example: a database server) goes down, using the messages and logs from the replication service, the state machine can be rebuilt from its initial state to its last functioning state.
In this architecture, the BookKeeper server takes care of storing write ahead logs, while the Hedwig server makes sure the replicated messages transfer safely across the network.
What is new in this release:
- This release includes a lot of improvements to the bookie on-disk performance, a new statistics framework, and protobuffer protocol support along with numerous bugfixes.
What is new in version 4.2.1:
- This release fixes a major performance bug in release 4.2.0. All users of BookKeeper 4.2.0 should upgrade immediately.
What is new in version 4.0.0:
- Read from open ledger.
- Add Fencing to Bookkeeper.
Requirements:
- Java 1.6 or higher
Comments not found