Incidents II: Scale & Growth

Traffic 10× and 100×, hot servers and hot shards, read/write saturation, replication lag, sharding and zero-downtime migrations, global latency, static content, big uploads and millions of jobs

About 110 minutes · 15 exam questions · 8 interview questions · 2 code labs

What you will learn

  1. Incident: traffic is 10× and still climbing
  2. Incident: planning for 100× — what breaks, in order
  3. Incident: one server is overloaded while its peers are idle
  4. Incident: read traffic is too high for the database
  5. Incident: replication lag — "my update disappeared"
  6. Incident: write traffic is too high for the primary
  7. Incident: the database needs horizontal scaling — do we shard, and how?
  8. Incident: one shard is overloaded (bad shard key / hot tenant)
  9. Incident: migrating data (resharding, new DB) with zero downtime
  10. Incident: global users see high latency
  11. Incident: static content and large uploads overload the app servers
  12. Incident: millions of background jobs

Previous: Incidents I: Slowness & Data-Layer Pain · Next: Incidents III: Async, Dependencies & Resilience