raft-consensus

README

raft-consensus

An implementation for raft consensus algorithm in Go

Goals

cluster.yaml

NODE (DB) (KAFKA) => receive some data from the client (SQL/REDIS command) => functions for processing this input and returning response back to the client => it should be always listening for new connections

RAFT NODE SERVER => receive some data fomr the client ([]byte) => replicate the data between RAFT NODEs

• RAFT library => include the raft settings so that there can be communication between this NODE and other NODES in the cluster => it should be able to participate in a clustered environment

raft-server1 --server "localhost:9091,localhost:9092"

raft-server2 raft-server3

type Store interface { GetLastUpdate() any Insert(data any) }

database empty database type calcState struct { students []string }

type state interface {

} raft.Start(stateMachine) raft.addLog() // raft.getlatestlog() // return term and the index stateMachine.Process() // commit

DS

• Kafka (STORES)

  • Doesn't process messages

• DB (PRODUCER, CONSUMER)

  • Process messages

raft call one function, that can perform operations on the state machine => create database postgres GOAL: Fault tolerance state_machine empty => [postgres]

=> create table table1 state_machine [postgres] => [postgres,table1]