✨ Add replication

[linkdd]

Scope

storage

Desired Behavior

At the moment, the solution is single-node only. We need to implement at least replication to avoid Single Point Of Failures in the infrastructure.

This can be achieved using the Raft protocol, which would replicate a history of commands to perform on the database (which would translate to BadgerDB transactions) to every node in the cluster.

See: https://github.com/hashicorp/raft