All challenges

Multi-AZ everything (four nines)

intermediate
Scenario & brief

A revenue-critical service must reach 99.99% availability ("four nines"). Today it runs two instances in two AZs with a single database — good enough for three-and-a-half nines, but not four.

Targets: 99.99% availability, critical durability, p99 ≤ 160 ms, under $1,600/mo.

Four nines means removing every single-AZ dependency: run the app across three AZs (with enough instances that losing one AZ still serves peak), and make the database Multi-AZ and add a read replica so the data tier also survives an AZ failure. A cache keeps the database comfortably in range.

4,000 rps peakp99 ≤ 160ms99.99% availdurability: criticalbudget $1,600/mo

App Servers

Compute

2instances

System health

Erupting · SLA breach

28

/ 100

Score

SLA not met yet

Monthly cost

$653

Budget $1,600/mo · within budget

Metrics

Capacity43
Availability90
Durability25
Cost efficiency84

Requirements

  • Peak capacity 3600 rps compute · 1700 rps db (need ≥ 4000 rps)
  • p99 latency ~194 ms (need ≤ 160 ms)
  • Availability 99.95% (need ≥ 99.99%)
  • Durability at risk (need redundancy + backups)
  • Budget $653/mo (need ≤ $1600/mo)

Advisor

  • Compute tops out at ~3600 rps but peak demand is 4000 rps — requests will queue.
  • The database is saturated at peak — add a cache to shed read load, or scale it up.
  • Enable automated backups to satisfy the durability requirement.

Discussion

Sign in to join the discussion.

No comments yet. Be the first to start the discussion.

For learning purposes only. Costs and capacities are illustrative, not live AWS prices. Not affiliated with or endorsed by Amazon Web Services.