Consensus: An Introduction to Raft

Transcript

1. Consensus An Introduction to Raft

2. con·sen·sus /kənˈsensəs/ Agreeing upon state across distributed processes even in

   the presence of failures.

3. Problem • Distributed System • Consistency • Partition tolerance

4. Solution • Quorum • Replicated State Machines

5. Consensus Data —

6. We are sacrificing Availability

7. Why not Paxos? • Difficult to understand • Not practical

   enough to implement

8. Raft A Practical Paxos

9. Components • Consensus Module • State Machine • Log

10. Consensus Module • Roles: Leader, Follower, and Candidate • Time

    is divided into Terms • Commands: RequestVote and AppendEntries

11. Leader Accept commands from clients, commit entries, and send heartbeats

    Follower Replicate state from leaders and vote for candidates Candidate Start and handle leader elections

12. Follower Candidate Leader Times out, Starts election Times out, Restarts

    election Wins election Discovers new leader, Steps down Discovers current leader or new leader, Steps down

13. Term Higher numbers are used to determine leaders and check

    log entries. The term is incremented each time an election is started. Any command with an old term is ignored.

14. Example Happy Log Entry

15. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [] Role: Follower Term: 1 Commit Index: 0 Log: [] Role: Follower Term: 1 Commit Index: 0 Log: []

16. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [] Role: Follower Term: 1 Commit Index: 0 Log: [] Leader receives command ˒

17. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [˒] Leader sends log entries to followers

18. A B C Role: Leader Term: 1 Commit Index: 1

    Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [˒] Majority of followers respond with success

19. A B C Role: Leader Term: 1 Commit Index: 1

    Log: [˒] Role: Follower Term: 1 Commit Index: 1 Log: [˒] Role: Follower Term: 1 Commit Index: 1 Log: [˒] Leader sends commit index to followers and responds to client

20. Example Sad Log Entry

21. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [] Role: Follower Term: 1 Commit Index: 0 Log: [] Role: Follower Term: 1 Commit Index: 0 Log: []

22. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [] Role: Follower Term: 1 Commit Index: 0 Log: [] Leader receives command ˒

23. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [] Leader sends log entries to followers

24. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [] Majority of followers do not respond

25. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [˒] Role: Follower Term: 1 Commit Index: 0 Log: [˒] Leader continues to retry log entry

26. Example Leader Failure

27. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [] Role: Follower Term: 1 Commit Index: 0 Log: [] Role: Follower Term: 1 Commit Index: 0 Log: [] D Role: Follower Term: 1 Commit Index: 0 Log: []

28. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [] Role: Follower Term: 1 Commit Index: 0 Log: [] Role: Follower Term: 1 Commit Index: 0 Log: [] Followers do not receive heartbeat D Role: Follower Term: 1 Commit Index: 0 Log: []

29. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [] Role: Candidate Term: 2 Commit Index: 0 Log: [] Role: Follower Term: 1 Commit Index: 0 Log: [] First follower to timeout becomes candidate D Role: Follower Term: 1 Commit Index: 0 Log: []

30. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [] Role: Candidate Term: 2 Commit Index: 0 Log: [] Role: Follower Term: 2 Commit Index: 0 Log: [] Candidate starts election and requests votes D Role: Follower Term: 2 Commit Index: 0 Log: []

31. A B C Role: Leader Term: 1 Commit Index: 0

    Log: [] Role: Leader Term: 2 Commit Index: 0 Log: [] Role: Follower Term: 2 Commit Index: 0 Log: [] Followers respond with votes D Role: Follower Term: 2 Commit Index: 0 Log: []

32. Extras • Log safety and compaction • Cluster changes

33. Real-life Application • Distributed lock server • Configuration management •

    Background job storage

34. Smart People • Raft Paper by Diego Ongaro and John

    Ousterhout • Raft Implementation • ThinkDistributed