Data Center Redundancy and Tiers Explained
Redundancy in data centers is critical for ensuring IT equipment is not impacted should there be a power disruption or equipment failure. The four data center tiers as classified by the Uptime Institute provide the international standard for data center performance, and higher tiers require higher levels of redundancy.
Data Center Tiers
The four data center tiers match the system availability needed for a particular business function and define criteria for maintenance, power, cooling, and fault capabilities. Each of the tiers includes the required components of all the tiers below it.
A Tier I data center is the most basic capacity level that must include a UPS, an area for IT systems, dedicated cooling equipment, and a generator. No redundancy is required. Tier II has a single path for power and cooling and adds redundant and backup components. A Tier III facility has multiple paths for power and cooling with redundant components to serve the critical environment. Maintenance and equipment replacement can be performed without taking the system offline. A Tier IV data center is completely fault tolerant with redundancy for every component. It has multiple independent, physically-isolated systems.
Data Center Redundancy
Different data centers provide different levels of power redundancy depending on the assigned tier and user needs. Redundancy levels commonly applied in data centers are N+1, 2N, and 2N+1, with “N” representing the amount of capacity needed to power, back up, or cool a facility at full IT load. If a facility is classified as N, it means there is zero redundancy built in.
N+1 adds an additional component to support a single failure or required maintenance, typically one unit for every four needed. So, for example, if an N+1 facility requires eight UPS units, it would have ten units. The same principle would hold true for a N+2 system in that the same eight-unit system would require twelve UPS units.
A 2N system is fully redundant, with a completely independent, mirrored system. An entire side or leg of the system can go offline with no interruption of service.
Highest data center reliability is provided by 2N+1, which combines the two levels above. This equates to a fully redundant, mirrored system plus one additional backup unit.
For more information on our data center capabilities, please contact Omni Instrumentation & Electrical Services, Inc. at 908-412-7130 (New Jersey corporate office) or 240-341-7915 (Maryland regional office).
Omni Instrumentation & Electrical Services, Inc., located in New Jersey and Maryland, is a premier instrumentation and electrical contractor, providing superior total care solutions since 1986. Services include instrument installation, tubing installation, instrument calibration, control loop testing, startup and commissioning, power and lighting, process control wiring, BMS wiring, telecommunications and data wiring, fire alarm wiring, security wiring, process network wiring, and control panel fabrication. Omni Instrumentation & Electrical Services, Inc., is certified in Foundation Fieldbus, Profibus, DeviceNet and ASI-Bus installation.