Jad Jebara, President and CEO of Hyperview, breaks down how to tackle a new era of data centre demand and power consumption.

Jad Jebara, President and CEO of Hyperview, breaks down how to tackle a new era of data centre demand and power consumption.  

Data centres – the engines that power our digital world – face a critical crossroads. The amount of data we generate is growing exponentially, fuelled by the rise of artificial intelligence (AI) and increasing amount of data being generated; meaning data centres need to consistently expand capacity to keep up – leading to an increase in energy consumption. 

A report by the International Energy Agency (IEA) predicts that data centres, along with AI and cryptocurrency mining, could double their electricity consumption by 2026. In 2022, these sectors used a staggering 460 terawatt-hours (TWh) of

 electricity globally. To put this into perspective, the average US household consumes about

10,500 kilowatts-hour (KWh) per year. 460 TWh is equal to 460,000,000,000 kWh, which is enough to power roughly 43.8 million US households for a year. By 2026, that number could balloon to over 1,000 TWh. That’s roughly the same amount of electricity used by Japan in a year.  

Data centres require significant power to run servers, cooling systems, and various other processes to ensure the entire infrastructure remains operational. However, high energy use of data centres translates to a large carbon footprint – further contributing to climate change and other environmental issues. 

Now is the time for data centre operators to find innovative ways to balance their need for ever-increasing power with the growing pressure to be environmentally conscious. Simple fixes like turning off unused lights fall short.  

So, what’s the solution? 

Data centre operators must consider more sophisticated, data-driven strategies. By implementing systems that capture detailed, real-time data, operators can track energy usage at a granular level; allowing them to identify areas for

 improvement and optimise energy allocation. It will also empower operators to make proactive decision-making, enabling them to anticipate and prevent issues before they arise.  

Measuring energy consumption by asset 

The current methods of measuring data centre energy use often focus on geographical location. This big-picture view, while helpful, doesn’t tell the whole story. It’s like looking at a forest from far away – you see the size and basic details but miss the intricacies of individual trees.   

Operators focusing solely on location will miss key details. These limited insights don’t reveal which servers are energy guzzlers, hindering targeted upgrades or replacements, and it doesn’t identify underutilised servers, opportunities for consolidation, or virtualisation to optimise resource allocation. 

To enable true sustainability, data centres need a closer look. Having the granular detail and being able to see the energy use for each piece of equipment – servers, cooling units, etc. is key to informed investment decisions for key stakeholders. With this knowledge, data centres can pinpoint areas for improvement and reduce overall energy consumption.  

The digital infrastructure maturity model 

The Digital Infrastructure Maturity Model, developed by the iMasons Climate Accord, has brought the industry together under a unified framework. It emphasises the need to measure the carbon impact stemming from power usage, equipment, and materials. 

By embracing this model, decision-makers can begin assessing their CO2 impact. This involves evaluating the carbon footprint of consumed power and the equipment sourced. In essence, an organisation’s CO2 output is the sum of emissions generated by power-consuming devices. This is greatly influenced by factors such as where the IT equipment is hosted, the power source utilised, optimisation of utilisation, the environmental impact of the supply chain and embodied CO2 in the facilities. 

Therefore, granular monitoring and reporting per asset become essential. This approach allows stakeholders to precisely identify underutilised assets based on various factors like age, type, function, and brand, as well as assess the impact from the supply chain. 

All-in-all, this is why actionable, detailed insights and continuous optimisation are important for sustainable operations. They empower decision-makers to improve the economic performance of infrastructure while simultaneously reducing its environmental footprint. 

How to drill down on energy usage 

Data Centre Infrastructure Management (DCIM) tools can help bridge this gap. These solutions offer a crystal-clear, real-time view of energy consumption – from the facility as-a-whole down to individual servers. This empowers operators to make smarter equipment choices, such as identifying and addressing servers that use a lot of power that do little work. 

This involves the essential collaboration between Colocation Operators and tenants, given their intricate interdependencies. Power Utilisation Effectiveness (PUE) for a facility hinges on how effectively tenants utilise their allocated power, constituting the

 sum of all tenants’ infrastructure, which the operator doesn’t directly control. Equally vital is Power Capacity Effectiveness (PCE), calculated as total power consumed divided by total power capacity installed. For instance, the Operator might build a 100 megawatts (MW) facility but if tenants only use 20 MW with a low PUE of 1.1, the PCE is 20%. Despite the low PUE, a low PCE prompts the need for additional facilities, impacting finances and the environment. As an industry, optimising both PUE and PCE at both operator and tenant levels, particularly in retail colocation facilities, is imperative. A low PCE indicates wastage of financial resources for tenants. 

Understanding the intricate interdependency between operator and tenant infrastructure is key. It operates as a cause-and-effect relationship, forming a continuous feedback loop. The more sustainable the tenant environment, the more sustainable the data centre becomes. Consequently, the financial performance and environmental impact of each asset directly influence those of the data centre. 

So, seeing exactly how much energy is being used at any given moment – real-time monitoring – enables operators to understand more about their biggest energy guzzlers. This allows them to optimise resources by consolidating workloads and using virtual servers, which ultimately cuts down on overall energy use. 

Balancing growth with sustainability 

Gartner reports a significant data centre spending surge driven by the AI boom, which requires additional amounts of vast power. Before rushing to buy new equipment, strategic planning for energy efficiency is vital. 

Data centres scaling for AI workloads must carefully consider the long-term operational costs associated with this additional power requirement. Striking a balance between growth and environmental responsibility is important. Here’s where comprehensive energy audits come in. These data-driven assessments identify areas for improvement within existing infrastructure, allowing data centres to optimise their existing resources before resorting to new equipment purchases. 

By embracing data-driven energy management strategies and prioritising efficiency through data-driven decisions, data centres can navigate the digital future responsibly. This approach ensures they have the necessary capacity to support technological advancements while minimising their environmental impact. 

The time to act is now 

Data centres need to prepare for the future. They will need to be both powerful and eco-friendly, especially with an increasing number of businesses adopting AI. This means completely changing how they manage energy. By going beyond basic measurements and using detailed data centre infrastructure management (DCIM) tools, data centre operators can find new ways to save energy and be more inherently innovative. 

However, they must be strategic before embarking on a tech spree. Operators should use data and insights to make informed decisions about what will benefit them and their stakeholders in terms of energy consumption and sustainability.   

As data centres chart their course forward, the adoption of advanced energy management strategies will undoubtedly emerge as a defining factor in shaping their success and sustainability in the years to come.  

Jad Jebara is President and CEO of Hyperview, a leading cloud-based data centre infrastructure management company.

  • Infrastructure & Cloud

Related Stories

We believe in a personal approach

By working closely with our customers at every step of the way we ensure that we capture the dedication, enthusiasm and passion which has driven change within their organisations and inspire others with motivational real-life stories.