What’s fuelling our data? ~ Improving energy efficiency in data centres ~

In 2018, Chinese data centres produced 99 million tonnes of carbon dioxide (CO2) — generating the equivalent environmental footprint of around 21 million cars. With global data traffic more than doubling every four years, it’s no surprise that the environmental impact of data centres is coming under scrutiny. Here, Simone Bruckner, managing director of power resistor manufacturer Cressall Resistors, investigates the efficiency of our data centres.

  • Friday, 7th February 2020 Posted 4 years ago in by Phil Alsop

Data technology is usually perceived as positive for the environment — think smart metres, predictive analytics and autonomous vehicles. In fact, even the notion of ‘The Cloud’ invites thoughts of crispness and clean air. However, the reality is that data centres consume a lot of energy.

 

As the uptake of cloud computing continues to increase at a rapid pace, more companies are relying on dedicated facilities to collect, store, process and distribute data. If data is going to continue to act as the powerhouse of the information revolution, evaluating its inefficiencies is crucial.

 

Power hungry

China’s data centre sector makes up eight per cent of the global market and is the second largest in the world. There are data centres everywhere, whirring away unseen across the globe. The biggest, covering over a million square feet, can consume as much power as a city of as many people.

 

The energy required to run a data centre can be broken down broadly into the power consumed by computing resources and that of supporting infrastructure, such as cooling systems. Typically, server rooms in data centres are cooled using classic ambient air-cooling with cold water-recirculation coolers. For high power applications, water-cooled racks are also used.

 

Brighter skies

Businesses are working to address the efficiency issue. In fact, there are a number of companies jostling for the title of greenest technology company. Both Apple and Google claim to run on 100 per cent renewable energy, while Microsoft announced that it is ahead of schedule to hit its target of 60 per cent renewable energy in its data centres by 2020. Renewables of choice for data centres include rooftop solar, wind, geothermal and waste heat reclamation.

 

The technology to improve data centre efficiency already exists. We’ve been harnessing the benefits of renewable technologies in a number of industries for decades, so there is no reason why data centres cannot benefit from the likes of solar power. However, the sheer mass of energy that a data centre requires means that ensuring the efficiency and continuity of renewable energy is crucial — data simply doesn’t have time for a cloudy day or broken panel.

 

When installing renewable energy systems onto data centres, operators will also need to put technology to avert potential inefficiencies into place. For example, when connecting new solar panels or disconnecting existing installations from the grid for maintenance, the panels continue generating electricity. This excess energy could damage equipment, so it must be safely dispersed.

 

To achieve this, load banks or dummy loads like those offered by Cressall should be added onto the solar panels to dissipate excess energy to help safely increase and maintain the renewable energy sources installed on a data centre. Load banks also help test a data centre’s air conditioning system — which is vital for keeping servers cool and functioning — to make sure it is working properly before the data centre is put into action.

 

While we cannot deny that the rise of big data presents a big energy challenge, they are the engines of the digital economy and it would be almost impossible for modern businesses to function without them.

 

The vast amount of energy that data centres require doesn’t mean that the cloud must return back to Earth with a bump. By making sure that renewable resources have the technology in place to provide more sustainable energy generation, we can continue to benefit from the wealth of data we create while minimising its impact on the environment.