Two nines — 99% availability
Two nines allows 87 hours and 36 minutes of downtime per year — nearly four full days. This tier is rarely acceptable for production systems with external users. It may be appropriate for internal tools, batch processing systems, or development environments where brief outages are tolerable.
At this tier, a single major incident lasting a weekend can consume the entire annual budget.
Three nines — 99.9% availability
Three nines (99.9%) is the most common SLA target for consumer web applications and cloud services. It allows 8 hours 45 minutes of downtime per year — roughly one major incident before the annual limit is reached.
Most teams find three nines achievable without dedicated reliability infrastructure. A single on-call engineer with good alerting can typically maintain 99.9% uptime.
Examples: Many SaaS products, e-commerce platforms, and internal business applications target three nines.
Four nines — 99.99% availability
Four nines allows only 52 minutes and 35 seconds of downtime per year. Achieving this requires active redundancy, automated failover, and disciplined change management. Planned maintenance must be zero-downtime by default.
At this level, a single unplanned outage lasting an hour can breach the annual SLA. Teams targeting four nines typically run multiple availability zones, use blue-green deployments, and have automated rollback on failed deployments.
Examples: Payment processing systems, core banking, healthcare record systems, and enterprise SaaS platforms.
Five nines — 99.999% availability
Five nines is the gold standard for mission-critical infrastructure. Only 5 minutes and 15 seconds of downtime per year is allowed — less than the time it takes most engineers to respond to a PagerDuty alert.
Achieving five nines requires geographic redundancy across multiple regions, automated failover within seconds, chaos engineering to validate resilience, and an on-call rotation with sub-minute response SLAs. At this level, even the deployment pipeline must be highly available.
Examples: Core telecommunications infrastructure, air traffic control systems, financial exchange platforms, and cloud provider control planes.
Six nines — 99.9999% availability
Six nines allows only 31.5 seconds of downtime per year. This is an extremely rare target outside of safety-critical systems. At this level, even planned maintenance must be executed without any service interruption, and recovery from hardware failure must be measured in milliseconds.
True six-nines availability requires active-active architectures across multiple geographic regions with no single points of failure at any layer of the stack.
The "nines" naming convention
The "nines" shorthand counts the number of 9s in the availability percentage. 99.9% has three 9s — three nines. 99.99% has four 9s — four nines. The naming convention breaks down above six nines, where it becomes impractical to express availability as a simple percentage.
Each additional nine reduces allowed downtime by approximately 90%. Going from three nines to four nines cuts annual allowed downtime from 8.75 hours to 52 minutes — a 10× reduction.
Composite availability
When a system depends on multiple components, the overall availability is the product of each component's availability. If service A has 99.9% availability and service B has 99.9% availability, the combined system has approximately 99.8% availability (0.999 × 0.999 = 0.998).
This is why distributed systems are harder to keep available than single-server systems — each dependency is a potential failure point that multiplies downtime risk.
SLA vs SLO — which tier to commit to?
External SLAs should always be set below internal SLO targets. If your engineering team targets 99.99% internally, committing to 99.99% externally leaves no buffer — any SLO breach immediately triggers a contractual violation.
A common practice is to set the external SLA one tier below the internal SLO. Target 99.99% internally, commit to 99.9% externally. The gap is your operational buffer.
See SLO vs SLA vs SLI for a full explanation of the difference.