Driving Efficiency: How IT Industry Emissions Propel Innovation in Software Development

  • 2024-05-07
  • Pauls Barkāns, Lead Solution Architect at Helmes Latvia

While much discussion revolves around transportation and aviation when it comes to environmental impact and CO2 emissions, far less attention is paid to the emissions generated by the IT industry because it doesn't seem as obvious. However, data from the International Energy Agency indicates that the IT sector accounts for approximately 5% of the world's CO2 emissions annually – more than aviation. Some experts express the opinion that by 2030, the IT industry could grow significantly and reach up to 20% of the total electricity consumption globally. Therefore, it is crucial to work on the creation of efficient software already.

To understand how to mitigate the IT industry's environmental impact, it is necessary first to identify which components are responsible for the largest emissions. Data centres and servers constitute a large portion (approximately 19%) of the total CO2 emissions from the IT sector. However, the lion's share (30%) consists of end-user devices - computers, printers, tablets, laptops, smartphones, smartwatches, etc. There are significantly more of these devices than servers, and their number and variety continue to grow rapidly, with users frequently replacing them with new ones. Another significant portion of emissions comes from IT services (15%) and network equipment (15%). The remaining portion (18%) consists of software, which can directly affect how much energy is consumed by the aforementioned devices and services and the amount of pollution they generate.

Awakening to Impact: IT Specialists Recognize the Consequences

One solution is sustainable software and green coding, which are relatively under-discussed topics, although not entirely new phenomena. In fact, in the early history of computers, when they were less powerful, programmers had to meticulously consider all algorithms, discard the unnecessary, devise and use various clever techniques and workarounds to fit within technological constraints. Creating efficient programs was a prerequisite for them to be installed and used at all.

Today, by developing software according to similar principles, we can say that in many ways, it aligns with the principles of sustainability and green coding. Many professionals understand these aspects and work daily to ensure that the software they create is fast, efficient, avoids unnecessary actions, and does not store unnecessary data. The difference between historical examples and modern times is that with the trends in technology and device development, specialists are beginning to realize the impact and consequences. At the same time, it should be noted that developing sustainable software is not solely the responsibility of programmers, as a large portion of software is created based on business requirements and needs. Sometimes it is necessary to take a step back and consider ways to make various business processes more efficient before programming and think about how to create these processes in the digital environment to work as optimally as possible, avoid generating unnecessary data, and leave as little impact on the environment as possible.

Data Volume's Environmental Toll: Infrastructure Costs and Impact Assessment

Data is still the new gold and can create extremely high value, but in many cases, the volume of data is growing without a specific, concrete, and understandable goal because processes for selecting the necessary data, processing it correctly, as well as archiving and deleting data, are not established. Sometimes organizations don't even have information about all the accumulated data or a clear vision of its application both at the moment and in the future. In such cases, the data volume increases, while data processing operations become longer and more cumbersome, requiring more equipment and resources for data processing. These additional resources not only have an impact on the environment but also create additional infrastructure and maintenance costs, which could often be avoided.

Awareness of existing data and analysis of business processes

What are the possible solutions to reduce the volume of stored data or perhaps store data in a more environmentally friendly way? It should be emphasized here that possible actions depend heavily on various factors - whether it is about creating a completely new system or improving the operation of an existing system. The most appropriate solutions also largely depend on the company's specific activities and business processes, the volume of existing data, the ability to adapt software to modern requirements, and so on. It is most often recommended to start with awareness of existing data and analysis of business processes; only then can decisions be made both about creating effective algorithms and other technological solutions.

If we are talking about creating a new system, it would be necessary to start with an assessment - what problems will the system solve, who will be its users, what are their needs. It is important to evaluate what will bring value to users and the company, what is really necessary, and what is insignificant or entirely unnecessary and will only incur additional costs in development, testing, deployment on servers, etc. From an architectural point of view, software should be built in a modular way, dividing it into interoperable components. This approach makes it easier to modify and replace systems in the future without affecting the overall operation of the software or completely shutting down a function if it becomes obsolete.

Meaningful and efficient solutions without unnecessary features

An important aspect that affects the overall result of the developed software is the methods used during development. For example, the "Helmes Latvia" team works with agile and lean methodologies, allowing them to build more meaningful and efficient solutions without unnecessary features, ultimately promoting cost savings in the long run, for example, considering the aforementioned electricity consumption. There are various technical solutions or approaches that programmers can use, regardless of the software, to promote lower electricity consumption. One example is data compression, which reduces the amount of data sent between devices at the network level. Another example is implementing data caching, which significantly reduces the use of server processor power for repeated uniform requests. If we use ready-made libraries in software development, we also initially compare the execution performance between multiple libraries to choose the most efficient option. We employ these and many other techniques when developing software for clients, but it is important to remember that these principles are best practices that complement rather than replace appropriate architecture choices and the creation of optimal various data processing processes.

Organizational Gains: Enhancing Reputation and Driving Business Success

When thinking about potential energy savings, it should be emphasized that optimization can be done in various places and stages. For example, at the International Conference on ICT for Sustainability, a study was presented using artificial intelligence to determine whether the content of a text is a spam message. The selected algorithm was able to identify unwanted content relatively accurately, and the study led to a new conclusion - by reducing the volume of the artificial intelligence data model by 80%, it operated with 75% less energy consumption, reducing precision by only 0.06%. We can easily imagine that in a system processing a large number of such messages, the potential energy savings with the correct algorithm use would be significant.

Finally, it should be noted that the benefits for organizations choosing sustainable software are diverse and largely depend on the company's goals. There are companies with high ethical and social responsibility standards that are keen on caring for the environment and reducing pollution. This undoubtedly also affects the company's reputation and increases brand value, as there has been an increasing interest from both the public and customers in these issues. For many companies, the goal is also to comply with the requirements of the European Union's Green Deal, as increasingly stringent regulatory requirements are planned in this area. Global research confirms that organizations that can reconcile business, technology, and sustainability goals achieve better business results. Therefore, by creating smart and thoughtful solutions, companies are undoubtedly winners in the long run.