Apache BookKeeper
The Apache BookKeeper subproject of ZooKeeper is made up of two major components:
- A distributed logging service called BookKeeper.
- A distributed topic based publish/subscribe service called HedWig.
We have implemented a JournalManager for Hadoop HDFS which allows you to run a high availability cluster using BookKeeper as shared storage. Details on this can be found in HDFS BookKeeperJournalManager.
is a replicated log service which can be used to build replicated state machines. A log contains a sequence of records which can be applied to a state machine. BookKeeper guarantees that each replica state machine will see all the same entries, in the same order.
Project Page
General information
- BookKeeper papers and presentations
- BookKeeper applications and usage
- BookKeeper roadmap
- HDFS BookKeeper Journal Manager
Architecture
(TBD)
- Bookie Storage Engine
- Client Request Flow
- Data placement and Replication
- Tailing Latency
- Ack quorum and Ensemble change
- Speculative reads
- Long poll reads
Developers
- Contributing Code Changes
- Code Style
- Git workflow TBD
Committers
Roadmap for work.