What should enterprises do, be wary of and consider when buying UPS equipment?

03 February 2022

Russell Bulley, senior application engineer, Vertiv

Russell Bulley, senior application engineer, Vertiv

Over the last year, we’ve seen the demand for data centres increase exponentially. As businesses become more reliant on digital services and home working, data centres became an even more essential part of how enterprises and society function. However, with a greater profile comes increased scrutiny.

What is a UPS?

UPS systems were designed to provide clean back-up power to data centres to any supply interruption, and to safeguard sensitive equipment against fluctuations or voltage spikes. This means key services can keep running either on battery or while backup generators kick in. The investment in a UPS means a guaranteed power business lifeline. In addition, innovative UPS’s may also provide a smart way to boost energy efficiency. This is because UPS batteries can store energy; so stored energy can be used when prices increase at peak times of day. This allows data centres to come off the grid at peak times and use UPS devices to become more cost-effective.

A business will need to work out which UPS is the right fit for its business. This comes down to knowing what you’re looking for and the functions that are applicable from the UPS. Part of this includes determining load size, resilience, budget, and equipment protection.

Determining the size of the load

Businesses can calculate the load size required by assessing which equipment warrants UPS backup. This is likely to be the IT equipment running critical loads. However, you may also want to consider whether other equipment, such as cooling systems or security infrastructure, also needs UPS support.

The next task is determining the power consumption (watts) of each device that needs to be connected to the UPS. This can usually be found from the equipment nameplate or manufacturer documentation. The sum of the power consumption of these devices will give you the required UPS kW capacity. For larger systems load power factor may be required.

Enterprises that determine their load correctly will find the most appropriate solution to efficiently power their critical systems. It is worth noting that UPS units come in two different formats: single phase and three phase. In general, loads less than 20kVA can safely use single phase UPSs, and larger loads tend to use a three phase UPS.

Keep on running

The next consideration is how much back up time individual data centres need should a power outage occur. This means understanding what their primary objective is for UPS backup power.
For example, many businesses may have a generator for extended backup power. These organisations will have a goal of keeping equipment running for the few minutes that it takes to transition over to the generator.

However, for others, the primary goal could be to have enough battery runtime to run for a period and then to safely shutdown servers and workstations using automatic safety software to avoid any data loss or corruption. This will require five to ten minutes of UPS battery autonomy. Alternatively, for critical applications such as mobile networks and Internet, you may want to have one hour of battery autonomy to be able to ride through even the most serious outages.

Determining primary goals for UPS power is an important step in understanding the run-time required and streamlining purchase decisions.

Planning for growth

Once a preferred runtime has been established; enterprises should consider how many devices need UPS support. The most successful enterprises will ensure that any single phase or rack mounted UPS’ have enough outlets to meet their immediate needs, leaving some room for growth. You may also use a power distribution unit (PDU) to provide additional outlets, being careful not to overload the UPS.

Special touches

Depending on your business requirements it may be worth considering special features that extend the functionality of your UPS system. Some common upgrades include:
• Remote Monitoring and Management – devices can be monitored and managed remotely. This feature is becoming more important given the huge growth in edge data centres with limited or no IT staff on site
• Lithium-ion Batteries – these batteries should last up to two-three times as long traditional lead-acid batteries yet weigh far less and take up less space. They also support faster charging and, many cases, lower the total cost of ownership (TCO) of a UPS due to longer life
• ECO-mode and Dynamic Online mode for Energy Savings – in an era where carbon footprint is front-of-mind, these could be key features for many. These features save energy by bypassing the voltage regulation or energy conversion processes of the UPS when the power input is of good quality.

While significant improvements have been made, downtime is still an issue in the data centre industry. According to Uptime’s Annual Survey, 69% of owners and operators surveyed in 2021 had some sort of outage in the past three years. Although this is a fall from 78% in 2020, roughly half of all data centre outages cause significant revenue, time, reputational damage and in this year’s report, 20% of outages were deemed severe or serious by the organisations that reported them. In addition, power remains the leading cause of major outages, responsible for 43% of outages in 2021. A relatively modest investment in a UPS is critical for any business mitigating against far more costly downtime and data loss.