Chinese Yellow Pages | Classifieds | Knowledge | Tax | IME

Raft consensus algorithm on distributed system

Raft:

paxos hard to understand, new consensus algorithm

consensus algorithm:  Leader elections, log replicate

https://github.com/etcd-io/etcd/blob/main/raft/README.md

 

This Raft library is stable and feature complete. As of 2016, it is the most widely used Raft library in production, serving tens of thousands clusters each day. It powers distributed systems such as etcd, Kubernetes, Docker Swarm, Cloud Foundry Diego, CockroachDB, TiDB, Project Calico, Flannel, Hyperledger and more.

 

Difference between Zookeeper

aft is a consensus algorithm/protocol, Apache Zookeeper is a product, a centralized service for maintaining configuration information, naming, providing distributed synchronization, and providing group services.

Zookeeper uses Zab as the broadcast protocol to propagate state updates between nodes in the ensemble.

So, if that makes sense, you should compare Raft against Zab or Apache Zookeeper against some other similar system like etcd.

Zookeeper development started in 2007, while the first drafts of the Raft algorithm were published in 2013. That said, Zookeeper’s Zab algorithm is actually fairly similar to Raft at a high level. It is not quite as simple as Raft, but it already works well and is supported by its own proof of correctness. Any simplicity advantage of replacing Zab with Raft in Zookeeper would be vastly outweighed by the years of work needed to implement, optimize, test, and mature a new algorithm.

 

References

 

 

Please rate this


Comments are closed here.