Data center tiers are a standardized classification system developed by the Uptime Institute. They are used to define the reliability and availability of a data center facility, primarily based on redundancy, power, cooling, and maintenance capabilities. Higher tiers represent greater levels of reliability and fault tolerance.
Understanding these tiers helps businesses choose a data center infrastructure that meets their specific needs for continuous operation and system uptime.
Data Center Tier Levels and Characteristics
Each tier builds upon the requirements of the tiers below it, increasing the complexity and redundancy of the infrastructure. Here’s a breakdown of the four tiers:
- Tier I: Basic Capacity
- Tier II: Redundant Capacity Components
- Tier III: Concurrently Maintainable
- Tier IV: Fault Tolerant
Let's look at what each tier entails and their associated uptime and downtime figures.
Tier I: Basic Capacity
A Tier I data center is the most fundamental level. It typically includes a single path for power and cooling, without redundant components. This means that any planned maintenance or unexpected failure can lead to downtime.
- Key Features:
- Single path for power and cooling.
- No redundant components (N).
- Susceptible to disruption from planned and unplanned events.
- Reliability: Offers the lowest level of reliability among the tiers.
- Uptime/Downtime:
- Uptime per year: 99.671%
- Downtime per year: <28.8 hours
Tier II: Redundant Capacity Components
Tier II facilities offer slightly improved reliability compared to Tier I by adding redundant capacity components. While it still relies on a single distribution path for power and cooling, critical equipment like power supplies or cooling units have backup capacity (N+1 redundancy) to handle some failures.
- Key Features:
- Single path for power and cooling.
- Includes redundant capacity components (N+1).
- Can withstand some equipment failures, but still vulnerable to path interruptions.
- Reliability: Better than Tier I due to redundant components.
- Uptime/Downtime:
- Uptime per year: 99.741%
- Downtime per year: <22 hours
Tier III: Concurrently Maintainable
Tier III data centers significantly enhance reliability. They have multiple independent paths for power and cooling, and redundant components (N+1). This design allows for concurrent maintenance, meaning that any component can be taken offline for maintenance or replacement without disrupting the operation of the IT equipment.
- Key Features:
- Multiple independent paths for power and cooling.
- Redundant components (N+1 or greater).
- Allows for concurrent maintenance of critical systems.
- Protected against disruption from planned maintenance.
- Reliability: Much higher reliability, designed for businesses that cannot tolerate downtime for maintenance.
- Uptime/Downtime:
- Uptime per year: 99.982%
- Downtime per year: <1.6 hours
Tier IV: Fault Tolerant
Tier IV is the highest level of data center reliability. These facilities have multiple independent active power and cooling distribution paths, as well as redundant components (typically 2N or 2N+1). A key characteristic is fault tolerance – the infrastructure is designed so that any single failure event anywhere in the facility (like a utility outage or a component failure) does not cause downtime.
- Key Features:
- Multiple active independent paths for power and cooling (2N or 2N+1).
- Redundant components (2N or 2N+1).
- Fault tolerant – designed to withstand a single failure without impact.
- Protected against virtually all planned and unplanned events.
- Reliability: The most reliable tier, suitable for mission-critical applications where downtime is unacceptable.
- Uptime/Downtime:
- Uptime per year: 99.995%
- Downtime per year: <26.3 minutes
Comparing Data Center Tiers
The differences in reliability are clearly reflected in the potential annual downtime for each tier, as summarized in the table below, based on the CoreSite reference:
| Tier | Uptime per year | Downtime per year |
|---|---|---|
| Tier I Basic Capacity | 99.671% | <28.8 hours |
| Tier II Redundant Capacity Components | 99.741% | <22 hours |
| Tier III Concurrently Maintainable | 99.982% | <1.6 hours |
| Tier IV Fault Tolerant | 99.995% | <26.3 minutes |
Understanding these tiers is crucial when evaluating data center providers, as the appropriate tier depends heavily on the business's operational needs, tolerance for downtime, and budget. Higher tiers require more infrastructure, leading to higher costs.
Reference: Breaking Down Data Center Tier Level Classifications - CoreSite
