Availability
ah-vay-luh-BIL-ih-tee
The percentage of time a system is operational and accessible to users.
Availability (uptime) measures the percentage of time a system is operational and accessible. It is expressed as a percentage: 99.9% availability means the system is down for no more than 8.76 hours per year. 99.99% means no more than 52.6 minutes per year.
Each additional nine is exponentially harder and more expensive to achieve. Going from 99% (3.65 days of downtime per year) to 99.9% (8.76 hours) requires redundancy, failover, and monitoring. Going from 99.9% to 99.99% (52 minutes) requires geographic redundancy, automatic failover, and extensive testing.
Availability targets should match business needs. A developer blog can tolerate 99% availability. A payment processing system needs 99.99%. Over-engineering availability wastes money. Under-engineering it loses customers. The target determines the architecture.
Examples
A company calculates monthly availability.
In March, the service was down for 20 minutes across two incidents. March has 44,640 minutes. Availability: (44,640 - 20) / 44,640 = 99.96%. The SLA promises 99.95%, so this month met the target with about 2 minutes of budget remaining.
A team achieves higher availability with redundancy.
The application runs on a single server: 99.5% availability. The team adds a second server behind a load balancer. If one fails, the other handles all traffic. Availability improves to 99.95% because both servers must be down simultaneously for an outage.
Availability math surprises a product manager.
The PM requests 99.999% availability (five nines). The engineering team explains: that is 5.26 minutes of total downtime per year. A single database failover takes 30 seconds. A single deploy takes 5 seconds. This target means fewer than 10 deploys per year if each causes even a blip.
In practice
Read more on the blog
Frequently asked questions
What does 'five nines' mean?
99.999% availability: no more than 5.26 minutes of downtime per year. This is the highest standard and is reserved for critical infrastructure (DNS, payment processing, emergency services). Most SaaS applications target three nines (99.9%) or four nines (99.99%).
How is availability different from reliability?
Availability is whether the system is up. Reliability is whether it works correctly when it is up. A system can be available (responding to requests) but unreliable (returning wrong data or intermittent errors). High reliability requires both availability and correctness.
Related terms
Service level agreement: a contractual commitment to specific performance and availability levels.
Service level objective: an internal reliability target for a service, like 99.9% availability or p99 latency under 200ms.
Service level indicator: a specific metric used to measure the reliability of a service, like latency or error rate.
An unplanned event that disrupts a service or degrades it below its expected quality, requiring a coordinated response.
Distributing incoming network traffic across multiple servers so no single server becomes a bottleneck.
Want the complete playbook?
Picks and Shovels is the definitive guide to developer marketing. Amazon #1 bestseller with practical strategies from 30 years of marketing to developers.