Skip to content

Systems Engineering Track

Consensus, Transactions and Resilience

Trace Paxos and Raft from durable state to quorum decisions, move work between systems that share no transaction, and stop one slow dependency from exhausting every thread.

Advanced · production systems
About 185 min reading
45 authored sections

Before you start: Sharding, Time and Coordination, or quorums, leases and fencing tokens in production.

Plan for 7–9 hours of hands-on work in addition to reading. This is an initial estimate; setup and extensions vary.

Open before purchase

A useful preview, not a teaser

Read these complete sections to judge the technical depth, teaching style, and architecture standard before buying.

Preview 1

Course Contract: What You Will Be Able to Operate

The third course in the systems sequence: agreement, cross-system work, and resilience. Failure first, mechanism second, vocabulary last.

Preview 2

Two Leaders, One Cluster

Each side of a partition can see itself. Neither can tell whether the other is dead or merely unreachable.

Preview 3

The Order That Was Paid and Never Announced

Two systems, no shared transaction, and a window that cannot be closed by ordering them differently.

Start the free preview

Complete curriculum

45 sections · 3 parts + resilience studio

Course orientation

  1. Course ContractFree
  2. Start Here If You Are NewCourse Pass
  3. The System So FarCourse Pass

Part 1 · Deciding Together

  1. Two Leaders, One ClusterFree
  2. Why Five, Not FourCourse Pass
  3. What Consensus GuaranteesCourse Pass
  4. Paxos Roles and StateCourse Pass
  5. Paxos in One Complete TraceCourse Pass
  6. From Paxos to Multi-PaxosCourse Pass
  7. Raft State and RPCsCourse Pass
  8. Raft: Terms and ElectionsCourse Pass
  9. One Raft Write, End to EndCourse Pass
  10. The Current-Term Commit RuleCourse Pass
  11. The Divergent LogCourse Pass
  12. Snapshots and CompactionCourse Pass
  13. Changing the Member SetCourse Pass
  14. The Deposed LeaderCourse Pass
  15. The Read PathCourse Pass
  16. Crash Faults and ByzantineCourse Pass
  17. Where This Already RunsCourse Pass

Part 2 · Work Across Systems

  1. The Dual WriteFree
  2. The Transactional OutboxCourse Pass
  3. Change Data CaptureCourse Pass
  4. Exactly-Once, HonestlyCourse Pass
  5. The InboxCourse Pass
  6. Two-Phase CommitCourse Pass
  7. SagasCourse Pass

Part 3 · Surviving Production

  1. Timeout BudgetsCourse Pass
  2. The Retry StormCourse Pass
  3. Hedged RequestsCourse Pass
  4. Circuit BreakersCourse Pass
  5. BulkheadsCourse Pass
  6. Rate LimitingCourse Pass
  7. BackpressureCourse Pass
  8. The Degraded ModeCourse Pass
  9. Proving ItCourse Pass

Resilience Studio

  1. Lab 1 · Paxos and RaftCourse Pass
  2. Lab 2 · OutboxCourse Pass
  3. Lab 3 · SaturationCourse Pass
  4. Challenge · Grant the VoteCourse Pass

Architect’s Desk

  1. ADR-003 · Order and EventCourse Pass
  2. When Not to Use ConsensusCourse Pass
  3. Review This ClusterCourse Pass
  4. AssessmentCourse Pass
  5. ReferencesCourse Pass