Find out exactly how much downtime is allowed under your SLA. Essential for DevOps, infrastructure and service desk teams setting availability targets.
Uptime & SLA Calculator — How Much Downtime Does Your SLA Allow?
Our free uptime and SLA calculator converts any uptime percentage into exact allowable downtime per day, week, month and year. Essential for DevOps engineers setting availability targets, IT managers negotiating service level agreements, infrastructure teams planning maintenance windows and cloud architects designing highly available systems.
The "Nines" — SLA Uptime Reference
- 99% (Two Nines) — 3.65 days downtime per year — unacceptable for most business systems
- 99.9% (Three Nines) — 8.76 hours per year — typical SMB SLA target
- 99.95% — 4.38 hours per year — typical cloud provider SLA (AWS, Azure)
- 99.99% (Four Nines) — 52.6 minutes per year — enterprise-grade, requires redundant systems
- 99.999% (Five Nines) — 5.26 minutes per year — telco-grade, requires active-active architecture with automated failover
How to Achieve High Availability
Achieving four or five nines requires eliminating every single point of failure. This typically involves: redundant servers in active-active or active-passive configuration, redundant network paths with automatic failover, redundant power with UPS and generator backup, database clustering with synchronous replication, load balancers with health checks, multi-region deployment for disaster recovery and automated monitoring with self-healing capabilities.
SLA Downtime in Plain Numbers
A 99.9% SLA sounds impressive, but that still allows 43.8 minutes of downtime per month. For an e-commerce site doing £10,000/hour in sales, that's £7,300 in lost revenue per month just from the allowed downtime window. Always calculate the business cost of downtime before agreeing to an SLA — and ensure your infrastructure can actually deliver what you're promising.
Planned vs Unplanned Downtime
SLA calculations must account for both planned maintenance (patching, upgrades, hardware replacement) and unplanned outages (failures, bugs, attacks). Blue-green deployments, rolling updates and canary releases allow you to deploy changes with zero downtime, protecting your SLA budget for genuine emergencies.