The anthropologist Steven Gonzalez Monserrate (2022) reports on his extensive ethnographic study of the cloud industry – detailing the operation of sprawling data centres, and the sometimes disorientating experience of working within them and/or living around them. Taking  into account wider connections to the worldwide infrastructure of telecommunications cabling and the ever-growing e-waste industry, Monserrate is able shine light on the material limitations and environmental consequences of our thirst for online data storage and computational processing.

The cloud comes at considerable environmental cost

Monserrate reminds us that the cloud industry is built upon a bewildering array of physical infrastructure, technical processes and computational logics – an assemblage that easily distracts from the environmental connotations of this industry. As he puts it: “the environmental costs of ubiquitous computing in modern life are obscured by the sheer complexity of infrastructures and supply chains involved in even the simplest of digital transactions” (p.2).

Nevertheless, by spending time immersed in the places where the cloud industry is physically located, Monserrate is able to paint a dispiriting picture of how our ever-expanding digital age comes at a considerable environmental cost.

At the heart of the cloud are sprawling data centres – huge metal and glass warehouses requiring massive amounts of power to function. These structures take up large tracts of land, suck up huge volumes of water to cool down the ranks of servers they contain, depend on constant air conditioning, and emit relentless levels of noise. Visiting one of these data centres soon brings home the “material flows of electricity, water, air, heat, metals, minerals, and rare earth elements that undergird our digital lives” (p.3).

Along the way Monserrate drops an array of alarming facts about these data centres. For example, the cloud industry now has a greater carbon footprint than the airline industry. In China, over 70 percent of data centre power consumption is drawn from coal. In some data centres as little as 6 percent of power consumption is used for actual data storage, with the vast majority used to power back-up systems, ‘fail-safe’ generators and servers – all redundant parts of the system ready to spring into action in the event of a breakdown.

Cloud industry pledges and promise might not be enough

Of course, Big Tech and the cloud industry are well aware of the poor public relations associated with the environmental impact of their operations, and so have begun to work hard to clean up the reputation of the data centre sector. We are now seeing grand pledges being made to move data centres away from a reliance on ‘dirty’ electricity sources, or to become ‘water positive’ by 2050. In Europe, a consortium of data centre companies has launched a ‘Climate Neutral Data Centre Pact’ – with a headline commitment to their facilities becoming ‘climate neutral’ by 2050.

Monserrate remains sceptical about the likelihood of any substantive shift from  ‘dirty’ to ‘clean’ data processing. He estimates that these Big Tech proposals might curb energy impact of the data centre industry by 20 percent, but reminds us that  there is no regulatory imperative for the remainder of the cloud industry to follow suit. Smaller data centre providers are likely to remain wedded to cheap ‘dirty’ power, and Big Tech companies do not have a good record of meeting previous pledges around carbon emissions and e-waste (Hogan 2018). As Monserrate (2022) concludes: 

“this dream of a haven for green data centres largely [seems] untenable to meet the computing and data storage demands of the wider world … Corporate pledges such as these, while laudable, are not enforceable, nor do they appear to be feasible given the explosive growth expected in data storage infrastructures over the next decade” (p.7, p.10).

Anticipating a world of data scarcity?

Of course, it might be that profit motives change, and the cloud industry somehow finds a way to continue its operations in cleaner, greener and leaner ways that make no negative impact on the environment in any way whatsoever. This is an industry that is now acutely aware of the need to “balance profitability with sustainability’ (Monserrate 2022, p.3).

However, Monserrate makes a good case for anticipating future scenarios where this is not the case – where our continued desire for more data eventually puts paid to our current assumptions of infinite capacity, power and expansion of  all things digital. Instead, we might anticipate a future of cascading systemic failures and data rationing – what Monserrate (2022, p.21) describes as a “world of bandwidth scarcity as the result of runaway desire, a culture of excess and unrestrained impulses”. 

The question that this then raises is straightforward enough – how might we steer ourselves towards a desirable version of this future? A future that does not descend in a digital ‘hunger games’, where the limited data and processing power that remains becomes the preserve of the elite, rich classes A future where the digital inequalities that many regions and demographics are currently enduring do not get ramped up to levels beyond our imagination …

**

References

Hogan, M. (2018).  Big data ecologies. Ephemera 18(3):631–57

Monserrate, S. (2022).  The cloud is material: on the environmental impacts of computation and data storage.  MIT case studies in social and ethical responsibilities of computing, Winter 2022.