Sammy Zoghlami, SVP EMEA at Nutanix, discusses data centres and the question of their ability to balance the demands of technology change with environmental concerns.
As pressure on data centres mounts due to energy consumption and a spotlight on how they impact net zero ambitions, organisations have to face up to reality. Despite incredible efforts to ‘green’ data centres in recent years, their ability to balance the demands of technology change with environmental concerns continues to be questioned.
In EMEA alone, data centres demand over 98 TWh of energy annually, equivalent to the consumption of an entire country like Belgium. And we are only at the start of an AI revolution. As IDC predicts, perhaps unsurprisingly, demand for AI is going to go through the roof over the next five years, so how data centres react and manage this demand is going to define the industry’s future role. Will it become an industry in constant flux, always trying to catch-up with data demands and compromising on sustainability concerns?
This is a trade-off. As businesses increasingly adopt AI, the infrastructure supporting it is under immense pressure. The question now is whether data centres can handle this surge sustainably? Most data centres are not built to cope with the high compute demands that AI workloads bring. The increased reliance on GPUs, necessary for processing AI tasks, leads to energy spikes and intensifies the cooling requirements, putting strain on existing resources. If left unaddressed, the gap between AI’s demands and data centre capabilities will widen, pushing the industry towards a breaking point.
Energy efficiency vs performance
One of the biggest challenges facing the data centre industry is the perceived trade-off between energy efficiency and performance. With AI, businesses often assume that to get the best performance, energy use must increase exponentially. But this doesn’t have to be the case. Hyper Converged Infrastructure (HCI) and next-gen cooling systems present viable solutions that offer both high performance and energy savings. By consolidating compute, storage, and networking, HCI reduces hardware needs and lowers energy consumption, making it a key enabler of sustainable AI operations.
As an Atlantic Ventures report Improving Sustainability in Data Centres 2024 reveals, in just seven years, modernising data centres with HCI could save up to 19 million tCO2e in the EMEA region, equivalent to the emissions of almost 4.1 million cars. It could also save €25 billion by 2030 from improved energy and operational efficiencies.
The point is that the technology already exists to address many of the challenges currently facing the data centre industry and as IT budgets are under pressure and electricity prices are soaring, energy efficient data centres and cloud operations are key levers for profitability and sustainability. And as the report confirms, hyperconverged infrastructure is at the forefront of transforming data centres into more energy-efficient and climate-friendly operations.
Regulatory pressure – No longer optional
As well as cost and contribution to net zero, there are also regulations to consider. The regulatory landscape is evolving rapidly, and data centres are under the microscope, especially as they now account for around 2% of the world’s energy consumption.
Sustainability initiatives like the European Union’s Green Deal and the Corporate Sustainability Reporting Directive (CSRD) are forcing companies to take action. Failing to meet energy efficiency and carbon reduction targets could lead to fines, reputational damage, and loss of business. The decision to reject Google’s Dublin data centre expansion is just the beginning. As governments crack down on high energy consumption and carbon emissions, organisations will need to proactively adopt green technologies or face harsh consequences.
While organisations are integrating AI into their operations, they must also align these advancements with their ESG (Environmental, Social, and Governance) strategies. The growing focus on ESG reporting means businesses need to demonstrate how their AI operations meet sustainability standards. This requires not only managing energy consumption but also using renewable energy sources and more sustainable infrastructure.
Forward-thinking companies are already exploring hybrid cloud models, where AI workloads are dynamically shifted to greener regions or times of the day when renewable energy is most abundant. But for many, making these adjustments is a daunting task, especially as AI demand continues to skyrocket. So, the future of data centres will depend on their ability to adapt.
The Atlantic Ventures report also found that switching from traditional 3-Tier architectures to HCI, can reduce energy consumption by more than 27% annually, helping companies cut both operational costs and emissions. Modernising data centres is now crucial, to cut the risks of facing a “doomsday scenario” where AI demand far outstrips available resources, leading to higher costs, more frequent service outages, and greater environmental degradation.
The solution lies in bold investments in energy-efficient technologies, smarter infrastructure, and renewable energy sources. As AI becomes more entrenched in our daily lives, the industry’s ability to balance its demands with sustainability will define its future – and the future of the planet.
Image Credit: Nutanix
