At the start of 2025, Mr Suricate is adopting greener practices for automated testing. Indeed, environmental protection is a priority for everyone in every field, so it's natural to ask: what about the world of automated testing? How can we make automated testing and web development more respectful of the resources available on our blue planet?
To answer this question: Marc Hage Chahine, QA Practice Manager at K-LAGAN. Host of the blog"La Taverne du Testeur " (which we strongly encourage you to read), and a regular at the JFTL, Marc's aim is to develop a qualitative vision of testing, taking into account sustainable quality and Green IT.
So let's answer this pressing question: what are the impacts of automation?
First of all, it's important to note that Green IT is not just a concept linked to the environment as such, but to the UN's sustainable development goals, taking into account the impact of the design and use of new technologies.
Marc reminds us of a study byADEME, published in January 2025, which states: "117 million tonnes of resources are used per year to produce and use digital equipment, i.e. 1.7 tonnes per French person per year", a trend that is rising sharply.
According to the report, the manufacturing phase accounts for most of the environmental impact, representing 60% of the carbon footprint, while the use phase accounts for around 40%. It should be noted that the latter figure has tended to increase since the recent democratization of artificial intelligence.
According to a study published by the U.S. Department of Energy, developed by Berkeley Lab, in three years' time, data centers could account for 6.7% to 12% of total electricity consumption in the United States. Between 2017 and 2023, data center electricity demand more than doubled as more AI-dedicated servers were deployed.
Obviously, we don't want to dramatize the situation or resign ourselves to it. The solution? Optimization!
Today, the aim should be for everyone to maximize optimization, avoid over-consumption and, at the same time, reduce the environmental impact of each and every one of us. As Marc explained to us, "we've been used to living with Moore's Law, with ever more powerful things, with the idea that we can always do more, with ever more graphics, ever faster and more efficient, and to living in this abundance, in this illusion of infinity".
Understanding its impact also means understanding that more and more is synonymous with over-consumption. AI is an excellent example of this. If it's not carried out within the right framework, doing AI because you simply want to do AI can be destructive for the environment. If you feel like doing AI, it's important to ask yourself: does what I want to do already exist in another, less energy-consuming form?
Is the use of AI really necessary in this case? AI is very similar to automated testing in this respect, even if the latter is less consuming. To find out more about this subject, take a look at theAFNOR standard, whose aim is to measure and learn how to reduce the environmental impact of AI.
Question internaute :
Can an AI with a resource usage more focused on the bare necessities, as Chinese and French AI do, considerably minimize the resource used?
For Marc, yes, this type of AI does allow you to do less, but you have to ask yourself the question of necessity. As explained above, if a simple Google search is enough to fill a need, AI may not be necessary (a Google search will always consume less than an AI search such as Chat GPT). SLMs(short language models), specialized models on a subject, unlike other general models, can be solutions.
To illustrate this difference, imagine that these models are athletes, one specialized in the 100m, while the other is more of a decathlete. If both are asked to run 100m, the former will have an advantage. On the other hand, if both are asked to hurdle, javelin or shot put, the latter will have the advantage, but will need much more energy.
The important thing to remember is that you have to make choices.
So, to sum up, Green IT is about the environment, people and sustainable quality. For more details, please visit La Taverne Du Testeur, which regularly deals with this subject.
We can then focus on the RGESN, the general eco-design standard for digital services. Once again, the aim is to limit our impact. The RGESN takes into account both tangible and intangible aspects, i.e. design. The reference framework is based on the following 9 themes:
This benchmark makes it possible to think about a wide range of uses, and indicates the ways in which we can really take action on our carbon footprint. Let's take the example of cell phones. Today, we're used to changing our phone after 2 and a half or even 3 years of use, compared with 18 months just a few years ago. But that's not enough: we should keep our phones for at least 5 years.
But why do we change phones so often? Marc talks about three main reasons:
RGESN also means thinking about the servers we use, as Marc tells us about his experience with his blog. With servers in the USA, La Taverne du Testeur didn't have the greenest servers it could have; the best would be to have its servers as close as possible, potentially powered by renewable energy.
According to ADEME, the impact on the environment is divided between terminals (50%), data centers (46%) and the network (4%). When we talk about data centers, we are of course concerned with the storage of our applications and their real needs. Storing data on 10 Tera or 1 Tera does not have the same impact on the environment.
Mr Suricate is a No-Code automated testing solution. We work mainly with major accounts, and position ourselves as the one-stop-shop for testing, i.e. we are able to carry out both production monitoring as well as bug detection in a test environment.
We also offer load shotsto obtain KPIs on route performance, as well as analyses of security practices. We also integrate analytical data tests, digital accessibility tests and Green IT assessments.
The Green IT dashboard enables us to assess Mr Suricate 's rating using the EcoIndex, a key environmental performance indicator. We are currently working on optimizing our site to improve this rating.Among the data reported, we analyze the number of requests and the complexity of the DOM, the latter being a determining factor in the EcoIndex: the more complex a DOM, the more difficult it is to display, and therefore the more energy it consumes.
At the same time, we worked with the K-LAGAN team of experts to develop a specific report for the RGESN (Référentiel Général d'Écoconception des Services Numériques), based on theENVI Score.
While theEcoIndex analyzes the impact of use, the RGESN focuses on design. Having both indicators allows for a more comprehensive and in-depth analysis.
The ENVI Score isbased on a number of criteria, such as architecture, content, UX/UI and so on. These scores are averaged to obtain theENVI Score, which constitutes a true audit. The results enable us toidentify areas for improvement and make appropriate recommendations.
This project is a co-development with K-LAGAN. We set up the KPIs, which are accompanied by the Green IT part, and K-LAGAN carries out a periodic audit to update your report, accessible directly from your Mr Suricate space.