Skip to main content

Demystifying cloud carbon emissions reporting

When it comes to decision making in IT, cost, reliability, security, and time to market have been the dominant factors for many years. These traditional factors are relatively straightforward to calculate, often based on the description written on the ‘tin’.

Matt Larder 180 x 180

Matt Larder

Head of Cloud, Softcat

When it comes to decision making in IT, cost, reliability, security, and time to market have been the dominant factors for many years. These traditional factors are relatively straightforward to calculate, often based on the description written on the ‘tin’.

However, incorporating sustainability and the impact of your decisions on carbon emissions, is fast becoming a new dominant factor. It’s still somewhat of a ‘dark art’ for three reasons, of which we’ll explore in this blog.

Incorporating sustainability into the Architectural process

This is potentially only relevant when designing new systems, or possibly when retrospectively reviewing existing architectures. In either case, the premise includes the introduction of a sustainability pillar into the architectural process.

Personally, I view this inclusion as not very complex. For example, an additional check and balance to ensure the recommended architecture is evaluated based on its cost, reliability, security, time to market and impact on an organisations carbon emissions. Moreover, many of the modern architectural frameworks already have a sustainability pillar built into them.

However, the practice is fundamentally more complicated, due to a lack of available data for the solution being architected.

For example, if an organisation is evaluating the replacement of end user devices, then this is potentially more straight forward as most vendors now publish carbon emissions data for the device (albeit with some working assumptions), which is the foundation of any architectural process in my experience.

This is vastly more complicated if you are tasked with a datacentre consolidation or migration to a new Cloud or SaaS platform. Similar to the previous example, some providers publish assumptive data or calculators for you to evaluate the options, but many still do not.

I think the answer here is a matter of time required for vendors and providers to provide more accessible data.

Visibility of carbon emissions

An example of additional commitments was unveiled by AWS at its annual Re:invent conference in November 2022, where by 2030 it plans to return more water to communities than it uses by 2030.  In contrast, I would also recommend watching this segment from the BBC, that provides additional insight against the subject of water positivity.

However, outside of the respective commitments, is the requirement to gain visibility of the carbon emissions and environmental impact of your cloud solutions. Aside from having visibility to ensure responsible decisions for the future of our planet, in January 2023, there was a tightening of the Corporate Sustainability Reporting Directive (CSRD), to bring more rules about the social and environmental information that companies have to report on, with the first companies needing to apply the new rules for the first time in financial year 2024.

Many cloud vendors and providers have well publicised mechanisms and dashboards to display your emissions, so there is a clear path to ensure adherence to the rules. But there is an equally well publicised ‘sting in the tail’ here, pertaining to data accuracy and accessibility as:

  • Much of the available data is typically only refreshed every 3-months
  • Data is only available at a geographic level (not at a regional level where your cloud solutions will exist)
  • And there is a level of ambiguity around how the emissions are calculated, with no common framework or bucketing of spend to calculate emissions.

Similar to the first point, one answer here is a matter of time required for vendors and providers to provide more accurate data. However, there has been a flurry of developments amongst start-ups and well-established cloud optimisation/governance software platforms, to incorporate a more accurate and real time approach to analysing cloud carbon emissions. Many of these solutions are built on Cloud Carbon Footprint, coupled with unique algorithms these businesses have built, to analyse your cloud spend, the region you operate in and services you use, to calculate your carbon emissions.

Many of you may be asking why such a third-party solution is required if the Cloud vendor or provider closes out their gaps and the answer is simple – they provide an abstraction layer to integrate emissions data from your multiple clouds, in oppose to you having to monitor and report on each Cloud independently.

Actioning of possible optimisations

Lastly, there is the need to report and action the insight you have gained from the previous point on visibility of carbon emissions. At this point, I refer to the next trend or buzzword in IT, GreenOps.

Taking a slight segway, many of you will be well versed in FinOps which is a Cloud financial management discipline and cultural practice, that enables organisations to get maximum business value by helping engineering, finance, technology and business teams, to collaborate on data-driven spending decisions.

Typically, FinOps requires a function within your Cloud Centre of Excellence (CCoE). This could be a dedicated function or set of people responsible for monitoring, reporting and actioning commercial governance, or the integration of this mindset into the various teams responsible for Cloud.

So conversely, GreenOps can be viewed as the sibling to FinOps, for a sustainability management discipline, associated function, within your CCoE, that is responsible for monitoring, reporting and actioning sustainability governance across your use of Cloud.

In my view there is two prevailing schools of thought around the GreenOps discipline; this needs either another function / set of people, or we see the merge of GreenOps and FinOps functions into one team. My personal view is the latter is the most appropriate.

In summary…

Hopefully the conclusions you draw from this are aligned to your existing overall approach to Cloud governance, leveraging a range of data points and design/build/operate principles, to ensure your Cloud solutions are reliable, cost effective, secure and sustainable.

If you have drawn this conclusion, then hopefully the incorporation of sustainability into your IT decision making process is not a million miles away for you, but perhaps requires time or some supplementary tools and process for you to achieve.

If you need any support with any of the topics mentioned, please get in touch to find out more info at