ENERGY EFFICIENT DATA CENTRES
ARTIFICIAL INTELLIGENCE: THE KEY TO FACILITATE MORE RESILIENT, EFFICIENT, DATA CENTRES
In addition to securing reliable electric power for data
centres, enhanced energy efficiency is essential to meeting Net Zero goals. David Pownall, vice president, Services, Schneider Electric UK and Ireland, explains how AI can help
T
he number of connected devices across the world is predicted to grow to a staggering
40 billion by 2030, with data centres forming the backbone for this expansion. Whilst technological advancements continue to boost economic growth and create a more hyperconnected society, the rapid pace of innovation is beginning to place a significant strain on data centres. With electricity set to account for about 50%
of global energy use by 2050, it’s clear that new ways to manage electrical assets, and ensure their resiliency, will be required. Securing reliable electric power is particularly crucial for data centres, where outages are estimated to cost over £200,000 per hour. Simultaneously, operators will need to enhance energy efficiency in these structures if they are to meet Net Zero goals. In Europe, major data centre operators have committed to achieving a power usage effectiveness ratio of 1.3 by 2030, down from the current average of 1.6 under the Climate Neutral Data Centre Pact. As the stakes rise, operators will need to ensure
that their electrical assets are dependable to minimise unplanned downtime. We’ll look at the ways in which artificial intelligence and data centre infrastructure management (DCIM) software are supporting data centres to become more resilient and sustainable.
AN INTUITIVE APPROACH TO COOLING The data centre industry globally produces about 11.8 million tonnes of waste electrical and electronic equipment, with overheated parts a significant contributor. Traditional power and cooling
18
optimisation technologies will need to evolve if they are to support the demands of higher density racks, which accommodate even greater amounts of computing power. Technologies such as liquid cooling, software- based cooling optimisation, and advanced airflow management are becoming increasingly popular, making it possible to maintain optimal temperatures whilst consuming less energy. With proper airflow management, operators can ensure that cool air is distributed evenly throughout the data centre, preventing hot spots and improving overall cooling efficiency.
AI INTEGRATIONS IN ACTION These technologies can be further enhanced with artificial intelligence (AI). When AI is integrated within an infrastructure management system, it collects and analyses data from thousands of sensors, monitoring variables such as temperature, humidity, server loads, airflow, and energy consumption. AI can also learn from external data sources, such as weather data. Instead of controlling cooling based on a fixed schedule, AI aggregates past data and predicted future insights to make adjustments in real time. This is a gamechanger for data centre operators looking to optimise their resources and prevent existing parts from overheating if a sudden shift
ENERGY & SUSTAINABILITY SOLUTIONS - Summer 2025
in weather, such as a heatwave, occurs. With tools that track energy usage, temperature, and performance metrics around the clock, operators can confidently allocate resources, as well as identify potential areas to optimise energy use. Many operators are already feeling the benefits: for example, Google’s AI implementation recently reduced cooling costs by 40%.
AI & AUTOMATION As well as predicting changes in temperature, AI algorithms can predict hardware failures and schedule maintenance before issues arise, to further reduce downtime and waste. This proactive approach ensures that equipment is maintained in optimal condition, extending its lifespan and improving its reliability. In some cases, proactive asset management has been proven to prevent critical asset failures by up to 60%, where maintenance visits are only needed every five years as opposed to every three. Artificial intelligence is also making a difference
for data centre staff by performing previously time-consuming manual tasks such as load balancing, backup management, and system updates, reducing the margin for human error. Automation of manual tasks enables operators to focus on more strategic activities which require human oversight. Into the future, AI is set to form the bedrock for the decarbonisation of complex data centres. AI-powered remote monitoring, cooling, and predictive maintenance will all play a vital role in facilitating resilient, future-ready data centres that the world can depend on.
Schneider Electric UK and Ireland
www.se.com/uk/en
David Pownall
www.essmag.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40
