What are data center tiers?
The tier system, popularised by the Uptime Institute, classifies facilities from Tier I to Tier IV by the redundancy of their power and cooling infrastructure and their ability to keep running during maintenance or faults. Many operators design to tier criteria without seeking formal certification.
Tier I to Tier IV at a glance
| Tier | Redundancy | Typical availability | Approx. downtime per year |
|---|---|---|---|
| Tier I | Basic capacity, single path | 99.671% | ~28.8 hours |
| Tier II | Redundant components, single path | 99.741% | ~22 hours |
| Tier III | Concurrently maintainable | 99.982% | ~1.6 hours |
| Tier IV | Fault tolerant | 99.995% | ~26 minutes |
Availability figures are the commonly cited values for each tier and are indicative, not guarantees.
Redundancy levels: N, N+1, 2N
N is the capacity needed to run the load. N+1 adds one spare component, so one failure or maintenance event does not cause an outage. 2N duplicates the whole system, so an entire path can be lost without downtime. 2N+1 adds a further spare on top.
How to choose the right tier
Match the tier to the cost of downtime. If an hour offline costs little, Tier II or an N+1 design may be enough. If downtime stops revenue or safety-critical services, Tier III or IV is justified.
- Estimate the cost per hour of downtime for your workloads
- Consider whether you can schedule maintenance windows
- Plan growth so you can move up in redundancy later
- Remember that operations and maintenance matter as much as design
Designing and building to a tier
Reaching a tier target requires coordinated design of power paths, cooling, controls and procedures, followed by integrated testing. See our data center construction service.