People are becoming more aware of their carbon footprint and while the usual suspects, aviation, personal cars and diet, are easily spottable, there is something less obvious and just as damaging – dark data.
Dark data refers to single use digital data, information, of any kind, that ends up in the cloud and is never used again. Some might consider the cloud is a magical space with unlimited storage capacity, but, in reality, all the data is stored in physical servers that take up space and consume energy.
Over half of the digital data that companies generate ends up being dark data, according to the World Economic Forum (WEF). It gets collected, processed and stored to only be used once. Old spreadsheets, old e-mails with who knows how many attachments. As for individuals, keeping 10 versions of the same picture on Google Photos or iCloud, for example, can have an impact.
The servers used to store this data consume a lot of energy, firstly just to keep them running, for temperature management, as they need to be kept cool to function properly. These data centres are responsible for 2.5% of human’s overall CO2 emissions, which is about the same as aviation, while digitalisation overall was attributed to 4% of the total emissions in 2020.
In our ever more connected lives, the data we now rely so much on also comes with a hidden carbon cost. Unsurprisingly, most of us don’t realise that our use of cloud storage means huge, power-hungry data centres are needed.
Chris Cartwright, Chair of the Digital Panel at the Institution of Engineering and Technology
The world is estimated to generate 97 zettabytes (97 trillion gigabytes) this year and 181 zettabytes by 2025. According to WEF’s calculations, a typical data-driven business such as insurance, retail or banking, with 100 employees, could generate 2,983 gigabytes of dark data per day. Yearly, this data’s carbon footprint is equivalent to 6 flights from London to New York. Overall, companies currently generate 1.3 exabytes (1.3 billion gigabytes) of dark data each year, having a carbon footprint of 3 million flights from London to New York.
This does not mean that digitalisation should be stopped. It does however mean that closer attention needs to be paid to digital data’s carbon footprint and measures should be taken to make current digital process more efficient, like shifting from single-use data and knowledge to its effective and efficient reuse.
