Do data centers need UPS? How to choose UPS?

Data center UPS (uninterruptible power supply), usually an online dual conversion UPS, is a very important equipment that can provide reliable power voltage for a period of time using a power source (usually a battery). Any power failure can have a destructive impact on mission critical computers, communication, and data, resulting in costly downtime. UPS helps with hardware protection, preventing data loss and damage, and providing continuity of service.

A、 How to deploy data center UPS?

In order to meet the high uptime requirements of data centers, UPS systems are typically deployed redundantly. There are three main UPS redundancy architectures, N+1, 2N, and 2 (N+1). N is the total UPS capacity required to handle the total load. In other words, N is the same as non redundancy.

N+1 uninterruptible power supply redundancy:

The N+1 UPS deployment provides the lowest reliability by adding components to support individual failures or requirements of that component. Here is an example to better understand N+1 UPS deployment. Imagine that you need to purchase 35 bananas for 35 students in a class and hope to buy an extra banana in case of unexpected shortage. 'N' represents the exact quantity of bananas you need (35), and the additional bananas are '+1'. Similarly, in an N+1 UPS redundant system, if the total load of the data center is 1000 kW and each UPS platform can handle 500 kW, three 500 kW data center UPS (N=1000 kW; N+1=1500 kW) are required.

2N uninterruptible power supply redundancy:

2N UPS redundancy is equivalent to a fully redundant data center architecture, divided into two ends (A end and B end) or mirrored systems. If two UPS are needed, there are two fully diversified and independent UPS to achieve flexibility. In 2N UPS redundancy, each side can handle 100% load capacity.

2 (N+1) Uninterruptible Power Supply Redundancy:

2 (N+1) UPS redundancy provides the highest reliability among the three in the data center. It is actually an N+1 system with double the required capacity plus additional capacity or redundancy. This redundancy level can tolerate multiple component failures and maintain N+1 redundancy in the event of a complete system shutdown. Due to being the highest level of UPS redundancy among N+1, 2N, and 2 (N+1), the cost of the 2 (N+1) architecture is always higher, regardless of initial component or operational costs.

B、 Factors to consider when purchasing data center UPS:

Here are ten main factors to consider when purchasing a data center UPS.

Uninterruptible power supply size: UPS systems come in various sizes and external specifications, such as desktop models, tower models, and rack mounted models. Desktop computer models have a compact structure and can be placed on a desk. The tower model stands upright on the ground or on a table/shelf. Rack mounted models are typically used for servers and network applications. For more detailed information, please refer to other articles published on this account.

Uninterruptible power supply capacity: measured in watts, UPS capacity is the power that the UPS system can provide. The higher the capacity, the more electronic devices or equipment it can support. The load (total power of multiple devices) should be calculated, and then the appropriate UPS should be selected. Usually, a UPS with an output power capacity 20-25% higher than the total wattage of the connected devices should be selected to help the UPS cope with fluctuations in power demand, leave room for other devices, and reduce the chance of overload. For more detailed information, please refer to this article on how to calculate the required UPS capacity?

Backup runtime: The runtime is the time during which the UPS system can use electricity to support connected devices during a power outage. UPS that accepts external battery packs can be used to extend operating time during power outages. The smaller the wattage load connected, the longer the battery's working time. The higher the wattage load, the shorter the running time.

Power supply voltage: Ensure that the UPS input plug/connector matches the socket of the input power supply. For example, the voltage in China is usually 220V, the voltage used to power servers and network equipment in North America is usually 120V or 210/240V, and Europe and Asia typically provide 230V power supply.

Number of sockets: Ensure that the number of sockets exceeds the current required quantity to ensure future growth space.

Data line surge protection: Choose UPS models with RJ11 jack, RJ45 jack, or coaxial connector to protect equipment from surges on connected phones, networks, or coaxial lines.

Pure sine wave output: The common power supply provides power in the form of sine wave alternating current. When UPS is in normal mode, it will transmit the same sine wave to the devices you are connected to. If the UPS switches to battery mode, it will generate sine wave or analog sine wave power to power your electronic devices. Some devices require pure sine wave power, such as computers with active power factor correction (PFC) power supplies. Determine whether to purchase a pure sine wave output UPS based on actual needs.

LCD control panel: Usually, it is recommended to use UPS with LCD display interface. It can display useful information such as voltage, frequency, backup time, etc., providing an intuitive view of UPS status and making system management easier.

Fault indicator light: LED indicator light informs users of potential hazardous wiring issues in the wall circuit.

Smart Slot/Smart Slot: Smart slots can use network management cards to customize UPS functionality. The optional network card allows comprehensive management through SNMP, web, SSH, or telnet. Smart slots are crucial for UPS in data centers to remotely monitor energy usage from anywhere and restart unresponsive devices.