Data center tiers refers to the Uptime Institute’s classification system of data center performance. Created over 25 years ago, data center tiers are still the standard in categorizing the level of service of a data center.
There are four data center tiers that each have criteria for maintenance, power, cooling, and fault capabilities. The tiers are progressive, meaning that each new tier must meet the requirements of the tiers below.
The data center tier levels are:
- Tier 1. A Tier 1 data center meets the basic capacity and infrastructure requirements to support IT in an office setting or more, including an uninterruptible power supply (UPS), a designated area for IT systems, dedicated cooling equipment that runs outside normal office hours, and a generator. Tier 1 data centers can help protect downtime from human error, but preventative maintenance and repairs require complete shutdown as there is no guarantee of power redundancy. Tier 1 facilities should have no more than 28.8 hours of downtime per year and at least 99.671% uptime. This tier is targeted to small businesses looking for a cost-effective solution that don’t have the requirements of the upper tiers.
- Tier 2. A Tier 2 data center includes some redundant power and cooling capacity components to provide better maintenance opportunities and safety from outages, including generators, energy storage, chillers, cooling units, UPS modules, pumps, heat rejection equipment, fuel tanks, and fuel cells. Tier 2 data centers have a should have no more than 22 hours of downtime per year and 99.741% uptime. Small- and medium-sized business clients often use this level as there are more guarantees of efficiency and redundancy, but still significantly lower costs than higher tiers. Tier 2 facilities can also accommodate more customers.
- Tier 3. Tier 3 data centers offer N+1 redundant distribution paths to serve the critical environment. These facilities do not require a complete shutdown to maintain or replace equipment, and any component can be shut down without impacting service. Tier 3 data centers should provide at least 72 hours of exclusive power to protect from power outages, have no more than 1.6 hours of downtime per year, and 99.982% uptime. This tier is for larger and more sophisticated customers that require dual power sources and redundant cooling.
- Tier 4. A Tier 4 data center includes multiple independent, isolated systems that serve as redundant capacity components and distribution paths to prevent an event from impacting several systems. Tier 4 facilities require all IT equipment to follow a fault-tolerant power design and continuous cooling to maintain a safe operating environment. At this tier, there are zero single points of failure, no more than 26.3 minutes of downtime per year, 99.995% uptime, 2N+1 fully redundant infrastructure, and 96 hours of exclusive power. Tier 4 data centers are for enterprise-level customers that can pay a premium to ensure the safe operation of their equipment in the event of nearly any threat.