Big Idea
Consensus is slow, but we need it! For example, IP assignment for servers and databases, replication schema, partitioning breakdown (what partition is located on which node). See Partitioning.
What is a coordinator service?
A coordination service is a key-value store that allows us to store this data in a reliable way. Some examples are: ZooKeeper (Zab), etcd (Raft)
- Put application configuration that require consensus into a coordination service
How do they work?
They are built on top of a distributed consensus layer like Raft. How to make sure reads are always latest?
- Read from leader - slow, high load on leader
- Read from multiple nodes, SYNC operation to keep reads linearizable.
Coordination Services Conclusions
Too slow to be used for application data, only key pieces of configuration for backend that need to be correct.
Built on top of consensus algorithms to maintain linearizability.