Data Centre Sustainability: if it were a hype cycle, where would we be?｜Business

The issue of climate change is widely regarded as the greatest issue facing humanity. Improving sustainability across all human endeavours contributes to addressing the issue. But is data centre sustainability living up to expectations? What might it look like if we think of it, and vendors’ efforts to mitigate it, in the context of a Gartner-esque technology hype cycle? How might we identify and label the gradients and inflection points? Where are we on the curve now, and what is its timescale? What will ensure data centre sustainability’s ongoing, permanent, relevance?

Talking about a (green) revolution

Most businesses have sustainability targets. Earlier this year 61% of IT managers and server procurement specialists told us that their organisation has overall sustainability and energy efficiency-related goals (with another 22% “working on it”). And, although the consumption of power by the world’s data centres is reckoned to exceed 1% of the world’s electricity production, a similar proportion (62%) of IT managers and procurement specialists told us that their IT function/department have formal sustainability and energy efficiency-related goals (with another 23% working on it). At the same time, geopolitical tensions continue to ensure that the cost of whatever energy is used continues to go only one way.

With the arrival of newer, more energy-hungry processors, requiring equally power-hungry cooling, sustainability – in the form of energy efficiency – remains high on the agenda of our customers whenever we initiate a conversation with them on the subject. It is not, however, raised spontaneously by customers with any great frequency. All of which seems rather out of kilter.

Data centre sustainability should be high on the agenda. Yet in May of this year, analyst group Gartner estimated that 75% of organisations will have implemented a data centre infrastructure sustainability program by 2027 (less than we might have presumed from our own survey results). Viewed cynically – i.e., from the opposite direction – we might say that only three-quarters of organisations will have done so, and it’s going to take them almost five years.

Sustainability in the data centre is, clearly, still a challenge to be faced down, but there are mixed messages about its degree of priority. Is it possible, then, that data centre sustainability has fallen victim to – whisper it quietly – hype? Out of curiosity, I wondered: might it follow the traditional hype cycle pattern identified by Gartner and since applied to many, many scenarios?

Scenario 1: we’re currently between ‘innovation trigger’ and ‘peak of inflated expectations’

We could be in this scenario. We know there is an issue, but the impacts of climate change have yet to directly and visibly affect data centre/procurement managers in most developed countries. Perhaps there is a subconscious assessment of ‘Yes, we use 1% of the world’s electricity production but, hey, that’s not really a lot, is it?’ going on. Despite the direction of travel for electricity consumption, as well as the rise of AI and digital generally, this scenario might look correct; it’s a plausible argument that we still have quite a while to go before we even think about whether our efforts to reduce energy consumption are working.

Points along this part of the curve might mark national and international targets resulting from, say, the Paris Agreement, or commitments made at various COP meetings. Perhaps points on this part of the cycle could note the ongoing work of the Al Gores and David Attenboroughs of the world, or the emergence of Greta Thunberg as a rallying point for public opinion?

If we accept that this is where we are, then what would mark the peak of inflated expectations? If we are in this scenario, then my own best guess is that the most likely trigger for what would signal the start of the descent into the trough of disillusionment might be some sort of disaster event. Such an occurrence might not be a single event, but a combination of events, for example:

The mindset that being environmentally sustainable is uneconomically unsustainable gains unstoppable traction
A climate-related disaster event – something huge that suddenly confronts developed countries with the impact of climate change and makes them think “All that sustainability effort, it was all for nothing!”
Regional wars or worse, trade relations have impacts over and above those foreseen

The consequence – tipping the cycle into its downward section, towards the trough of disillusionment – simply a higher immediate priority: [business] survival.

Scenario 2: we’re at the peak of inflated expectations

Are we already in this scenario; has this already happened? Instead of prioritising the climate to try and fix the underlying cause, are organisations just prioritising their economic survival? Consider the current, difficult economic environment: there are regional conflicts and geopolitical tensions everywhere you look, and difficult trade relations between some of the world’s largest economies. It’s surely a possibility that organisations have already reassessed.

Scenario 3: we’re already in the trough of disillusionment (or on our way to it)

Perhaps the reason customers generally don’t instigate conversations about sustainability metrics is because they are already in the trough of disillusionment? It’s an arguable theory: humanity has a history of successfully adapting (which might make “Adapting to survive” a suitable label for the downslope of the cycle into the trough). If we are at this point, how will the curve drag itself out of the trough of disillusionment and onto the plateau of productivity?

Might I posit that the answer to this question is some variation of “legislation”? Again, it’s arguable that we are already emerging from the trough (subject to identifying and labelling the previous inflection points). Right now, what's largely driving the push for sustainability is not users, or the buyers supporting those users. More than either of these, it’s legislation from governments that’s forcing the hands of buyers, or it’s the efforts of vendors trying to stay in line with legislation. As we get closer and closer to the dates regarded as the moment tipping points will occur, legislation is likely to become ever more pervasive and ever more stringent. As well as using sticks, governments may also come up with carrots, devising creative ways to incentivise businesses to go ever greener, including in the data centre.

A second marker for the bottom of the trough of disillusionment might be the continued occurrence of climate-related disaster events – causing exactly the opposite effect that they had at the peak of inflated expectations. Eventually, these occurrences might force sustainability to be reassessed as the highest priority for organisations, forcing them to do whatever they can do to be sustainable.

A third possible trigger for exiting the trough is new technologies. Do I know what these might be? No! But does technology ever stand still? Also no. We're unlikely to eliminate completely carbon dioxide as a byproduct of energy production. But we might find ways to convert that CO2 back into something usable. Experiments are already underway to remove carbon dioxide from the atmosphere, and to combine it again with hydrogen to create a new form of aviation fuel. Humans are ingenious, and doubtless there are many better examples now and to come.

Most likely, of course, a combination of factors will be involved in data centre sustainability dragging itself out of the trough of disillusionment, and onto the plateau of productivity.

Scenario 4: the issue becomes obsolete

It’s sometimes overlooked that, after the plateau of productivity, the hype wave tails off, falling down and right, back to the graph’s horizontal axis. This could be the result of a technology becoming obsolete, for example. What might cause the obsolescence of the sustainability issue?

Much as the emergence of new technologies could trigger the cycle to turn upwards again onto the plateau of productivity, the emergence of even more advanced technologies in the future could eventually render obsolete (or of negligible value) the whole concept of data centre sustainability. Two possibilities spring to mind, of course: the energy required to run data centres (and the rest of the planet) is sourced from entirely renewable sources, or effective (and cost-effective) means are found to capture and contain carbon dioxide. Beyond that, your guess is as good as mine.

Data centre sustainability matters. Thinking of it as a hype cycle is just for fun

The issue of climate change is widely regarded as the greatest issue facing humanity, and that improving the sustainability of human activities contributes to mitigating the issue. In casually wondering whether data centre sustainability may be following a hype cycle – and, if it is, where we might be in it – I am certainly not trivialising it. ASUS is heavily invested and a global leader in sustainability – both in respect of its own operations and the products we bring to market. We take it very seriously.

In terms of where we might be on a hype cycle for data centre sustainability… every point of view is possible (including that it doesn’t and won’t follow any such pattern). If it does follow the wave pattern, it’s possible to argue for us being anywhere on it. It’s also possible that different organisations are at different points on the graph at the same time, interpreting inflection points differently, to the extent that those inflection points apply to them. Ultimately, we might only find out years from now where we were in the mid-2020s. But if this article just keeps us thinking about the importance of data centre sustainability – at a time of many competing and even conflicting priorities – it will have been well worth the effort.

Hype Cycle™ is a trademark of Gartner®, Inc. It is a methodology that provides a view of how a technology or application will evolve over time, providing a sound source of insight to manage its deployment within the context of specific business goals. Gartner has not been consulted on endorsed or contributed in any way to this article.