Vertex

Replication and read scaling

Replicas give you read capacity and failover. They also give you a window where a follower answers with data that is slightly out of date.

Lesson 2.25m 2sIntermediate11.7k students

Chapters

5:02
  1. 0:00Database replication strategies for distributed systems
  2. 2:59Multileader strategy mitigates leader failure
  3. 3:43Leaderless replication Clever methods for managing chaos
  4. 4:41Choosing the right exponent strategy

What this lesson covers

  • Distinguish synchronous from asynchronous replication
  • Route reads to replicas without breaking read-your-writes
  • Plan and rehearse failover

This lesson sits in Data at Scale, part of System Design Foundations. It assumes what came before it and leads directly into the next lesson in the module.

In this lesson you will

  • Distinguish synchronous from asynchronous replication
  • Route reads to replicas without breaking read-your-writes
  • Plan and rehearse failover

Resources

Previous lessonSharding and partitioning2.1 · 6m 2sNext lessonConsistency models and CAP2.3 · 6m 48s