SUSTAINABILITY FOCUS 16

Sustainability IT and Circular IT in the light of COP26 By Betsy Dellinger, Senior Vice President and General Counsel, Park Place Technologies

  • 2 years ago Posted in

What did CTOs and CIOs learn in the immediate aftermath of COP26 that would force ICT to the centre stage of sustainable drives for cleaner technology adoption that would reduce the consumption overhead? Did any of the sessions pave the way for ICT to become pioneers in the delivery of sustainable, circular IT, or are we to continue as the consuming villains as pretty much everything ICT delivers is driven by electricity usage?

The top-line emission numbers discussed at COP26 were staggering and outlined the scale of the problem globally. The International Energy Agency (IEA) highlighted the main sources of global carbon emissions as being electricity, industry, and transportation. The electricity sector accounts for 40% of carbon emissions. Of that, current analysis shows that the ICT sector consumes 4% of global electricity used. And, left unchecked, such is the exploding growth of data compute requirements, analysts have predicted consumption by data centres alone will rise to a threshold of 10% by 2030. Overall, one of the five aims of COP26 was to dramatically accelerate the push to limit global warming to 1.5 degrees. To achieve that, analysis shows that the world needs to cut emissions by half and reach net-zero carbon emissions by 2050. In our industry, achieving carbon neutrality in IT has to come from both low-carbon electricity generation and reducing consumption. And that starts with the biggest guzzlers of all, the data centre.

In fairness, data centres have made some of the most significant gains toward environmental sustainability in recent years. Historically most data centres were anything but green, largely because most of the compute processing was fuelled by fossil fuel generated electricity. Being a high consumer is a difficult image to shake-off. Because at the heart of the problem is the world’s incessant demands for increased workload compute and data processing. With no option to cut growth of data, IT Leads face the unenviable challenge of offsetting exponential data processing needs with detailed ongoing sustainability programmes.

One such approach to cutting the carbon footprint is an ongoing reduce and recycle strategy. Here ICT leads operate with increased fortitude in their hardware lifecycle strategies and by adopting dramatically different attitudes to repurposing. This involves fast acceptance of further sustainable end-of-life management initiatives discussed in COP26; reducing consumption; seeking reuse and repurpose drives across estates, and when waste occurs, to positively harness by-products in a sustained push toward greater circular economy efforts.

Let’s first examine reducing inbound energy consumption on-premise through selection of optimal use of hardware and software assets in the estate, as cutting energy use remains the top priority. Reduction of consumption starts with understanding exactly what is drawing and using power from within the estate. This is asset discovery but with reduce power-draw top of mind. It is an essential first step to allow organisations to have a realistic benchmark from which to measure reductions in carbon footprint and achieve a realistic framework for accountability. This may sound like a high barrier to entry, as estates are incredibly complex, dynamic, and dispersed. To keep focus on the overall sustainability goals, it’s often better to pass this to a qualified third party who can automate the discovery and ongoing tracking and audit process, globally. In any sizeable estate, the results will quickly reveal immediate inefficiencies that can be tackled - flagging inactive or under-utilised servers. Results are individual, but as a general rule, users detect that up to 20% of servers draw power without doing any useful work, straight off the bat! With an asset discovery designed to highlight environmental sustainability, conclusions will allow organisations to both reduce needy power assets and to identify which assets to repurpose.

Stretching end-of-service-life (EoSL) hardware dates out of their stated manufacturer timeframes is equally compelling for increased sustainability, often layering software to optimise assets’ credible working lifespan. This isn’t a new concept. End of life has never really entailed immediate terminality or vulnerabilities that IT vendors mention; and stretching lifecycles without impact to digital gains is a great way to increase sustainability fast. At Park Place Technologies, we suggest an immediate gain is to question whether replacement for servers and storage that fall outside of EoSL is essential. A lifecycle survey conducted by a leading analyst firm highlighted that normal server usage can span between up to 5 years, and that 47% of all organisations credibly extended server lifespan for four years or more by applying the 80/20 rule. (If 20% of your applications supported by the hardware are deemed Tier One critical to business functions, then only these apps deserve access to latest generation hardware refreshes). The remaining 80% can be readily stretched, taking advantage of an extended warranty programmes delivered by a reputable Third-Party Maintenance provider (TPM). Deploying pre-owned hardware can also be explored through TPMs, which can give additional years of useful life from certified previous generation servers backed by lifelong guarantees on authenticity and assurances on firmware updates. If data centre leads lessen their dependance on latest generation hardware and re-purpose with confidence, then the knock-on impact in increasing sustainability efforts and reducing landfill will surely follow.

What about cloud, managed service users and specialized off-site data centre usage that includes co-location of private servers hosted in remote data centres, does that cut consumption or does it

simply shift the burden and the utilisation overhead to another facility? In some ways yes, but most cloud and remote data centres are modern, purpose built and designed from the ground up to take full advantage of the massive gains that have been made in creating more carbon friendly facilities so that the building infrastructure itself is highly efficient. Before making any switch to off-site, IT Leads embark on an in-depth review of the carbon credentials of remote data centres to lower their own carbon baseline. This would include determining increasing dependencies on renewable energy sources such as wind and solar, or in using environmentally friendly ways of creating electricity such as using lignite.

That takes us onto the massive waste overhead that data centres carry and what initiatives can be applied there, both in terms of reducing e-waste (electronic waste) with the discard of any electronic devices, and increasing efforts for effectively dealing with output waste through ongoing functions such as cooling.

Electronic waste remains particularly concerning. In November 2020, the Environment Audit Committee (EAC) highlighted that the UK was one of the worst offenders at shipping e-waste to developing countries for processing and disposal – at around 40% of total nationwide e-waste figures. Hardware scrap components such as CPUs contain potentially harmful materials such as lead, cadmium and beryllium, the recycling of which carries health and contamination risks in shipping, handling, and processing. There are no short cuts to minimising a data centre’s waste electrical and electronic equipment (WEEE) overhead, it’s down to reducing the levels of commodity hardware across the estate and planning with specialist partners on how to increase usability of existing resources. As more governments are now holding IT departments accountable to recycling and reuse targets, specialist providers should be sought who bring automation and intelligence into environmental infrastructure planning, setting objectives, proving adherence, and saving new purchases.

What about harnessing outbound waste from massive data centres for good? Data centres in their processing operations create lots of heat from cooling that is expelled back into the atmosphere. How can that be diverted in effective waste heat utilisation programmes known as Energy Reuse Effectiveness (ERE)? Here, using ERE provides data centres with the unique opportunity to switch from mass energy users to mass energy suppliers using the by-products of server cooling. In the UK, when the government launched its net zero strategy on the back of COP26 recently, it announced ambitious nationwide plans that would view data centres over a certain size as prime sources of ‘waste heat reutilisation sites’ to be tapped into by domestic and commercial users alike through the creation of heat networks from data centre waste outputs as a low carbon source of heat. Building the heat networks and pipe infrastructure has already started with serious decarbonisation results expected by 2025.

Using initiatives like these will see a serious reversal of fortunes for meeting carbon targets and achieving net zero overhead for data centres as ICT ramps up fast to becoming pioneers in the supply of sustainable solutions.

By David de Santiago, Group AI & Digital Services Director at OCS.
By Krishna Sai, Senior VP of Technology and Engineering.
By Danny Lopez, CEO of Glasswall.
By Oz Olivo, VP, Product Management at Inrupt.
By Jason Beckett, Head of Technical Sales, Hitachi Vantara.
By Thomas Kiessling, CTO Siemens Smart Infrastructure & Gerhard Kress, SVP Xcelerator Portfolio...
By Dael Williamson, Chief Technology Officer EMEA at Databricks.