Michael Akinla, Business Manager Central Europe North at Panduit EMEA, explores the role UPS plays in sustaining the expansion of the data centre sector amid increasing worldwide energy demands and evolving power grid infrastructures.
The pressure on energy networks to sustain the expansion of the data centre sector and IT-dependent enterprise has intensified due to the growing worldwide demand for energy and the lack of investment in power generation and transmission grid infrastructure.
This is made worse by our increasing reliance on renewable energy sources. While energy matching is a typical solution today, using renewable energy necessitates procuring power from many providers, including solar and wind, which are not currently guaranteed to be available continuously.
Power outages can have disastrous effects on companies, and according to the Uptime Institute, this includes between 10 and 20 high-profile IT outages per year that cause serious financial loss. Although there has been a reduction in outages, still 55% of operators that responded to the 2023 Uptime Institute data centre survey, reported an outage in the last three-year period. As the demands on ageing infrastructure increase major grid failures will occur even in nations with secure electrical infrastructures, such as the UK.
One of the most important parts of any setup where maintaining constant and consistent electrical power to IT equipment is essential is the UPS. MarketWatch projects that the uninterruptible power supply industry will reach $13 billion by 2025, growing at a compound annual growth rate of 3.98%.
Keeping IT systems online
Key components for the successful operation of IT systems are battery UPSs, diesel generators, and backup electrical generation or storage systems. When an outage occurs the backup generator power is not immediately available, and it is usually the UPS that provides the facility with the bridging power needed to run critical systems to maintain data or processes, or allow a structured shutdown of essential systems.
There is a vital window of time which is measured in milliseconds between the loss of utility power and the IT load switching to the UPS. Any power outage lasting longer than 20 milliseconds is likely to cause an IT system crash. If the outage lasts up to 60 seconds, it can also cause a protracted ITE restart procedure, which might have a major impact on data centre operations and customer applications. Long-term outages may also result in financial penalties from clients and harm to one’s reputation.
Choosing the right UPS
UPS components and technology are improving, which is changing the range of capabilities available. The necessity for backup energy is also changing due to the rapid development of faster processors and server storage. The UPS’ capabilities will be determined by the IT applications it supports, the customers’ risk tolerance, and the resilience of the applications. Therefore, before choosing the UPS to support this, the data centre engineering/operational team or customer must precisely identify their needs.
For many hyperscale data centres, the UPS chosen is required to maintain the energy supply to the supported system for as little as one to two minutes, while colocation sites normally need five minutes of power. However, the UPS runtime is usually between 10 and 15 minutes in the financial industry, where data is mission-critical and even a small number of dropped trades could cost many hundreds of thousands of pounds. The longer the duration of power needed, the higher the initial and ongoing expenditure in the number of available UPS to any specific customer.
On the other hand, over-purchasing UPS with long runtimes in situations when backup generators are easily accessible may result in extra ongoing costs and needless capital expenditures that can impact profitability.
Choosing UPSs that are appropriate for the IT load is crucial. ITE is the obvious target for UPS with its essential workloads, and taking a wider view, as faster processors increase server heat output, UPS for the cooling systems that supply those servers is also becoming a crucial load. Keeping that in mind, the UPS unity power factor is another crucial component of the procurement cycle. The unity power factor of the latest UPS is 1, therefore today, a customer’s 100kW IT load could be covered by 5x 20kVA UPSs or a single 100kVA UPS, depending on the preferred configuration.
However, if installing older model UPS this could have a bearing on its unity power factor, which could be less than 1, and that would impact the UPS configuration needed to sustain the critical load.
Bringing together the old and the new
When merging older UPS systems in an environment that is moving to higher power-rated ITE racks, modular UPS components are a solution. The design and components of the latest systems can lower costs and increase energy efficiency by matching the rack kW with improved UPS kVA, which is crucial to optimise battery life.
Lithium-ion batteries, which are now firmly established in the market, will provide consumers with increased capacities. Li-ion battery UPSs are gaining market share in edge data centres and off-site locations due to their longer lifecycles, lighter weight, smaller footprint, and lower cooling requirements, compared to lead-acid batteries. Li-ion batteries’ ability to provide twice as much energy in smaller devices at a factor of two to three, along with cooler operation that eliminates the need for specialised cooling systems, raises the possibility of doing away with the need for separate battery rooms and increasing space utilisation.
Many new device management and safety features are being incorporated as technology advances providing detailed data on battery health and connectivity. Increased connectivity options, as well as the ability to disseminate that data across the power and data centre network, are also available.
The Li-ion UPS can be integrated into pre-configured data centres, enabling customers to order using a single SKU to include the entire UPS system, have it shipped to any location, install it easily, connect it to power, and start using it. These expanded capabilities are supported by the intelligent power distribution units (iPDUs) in the racks, which offer additional communication and data management capabilities.
The key to the data centre’s future
UPSs offer higher efficiency, dependable power protection, and backup power for IT systems and other vital equipment. When necessary, modular systems enable hot swapping, which paves the way for quicker maintenance, and the replacement of outdated or malfunctioning individual units to guarantee maximum capacity and instantaneous power availability.
The latest systems offer superior electrical performance, intelligent battery management, enhanced intelligent monitoring, secure networking services, and a long lifespan for lithium batteries to satisfy the expanding power demands of data centre, enterprise, and edge IT equipment. Support for administration, monitoring, control, and alerting throughout the wider environment, including power chain, environmental, cooling, security, IT assets (Physical & Logical), and connectivity infrastructure, is provided by integration with cloud-based DCIM solutions.
The clear objective is to keep vital assets from going offline. The constant evolution of technology makes it more difficult to minimise disruptions, and that’s why UPSs continue to be an essential component of data centre best practices.
