There is an interlink between resource extraction and digitalization, particularly in information and communication technologies (ICT) systems like data centers. Regenerative design can serve as a solution to develop eco-friendly and efficient systems that actively enhance natural ecosystems and restore their own sources of energy and materials.
- The principles of regenerative design include: safe and healthy materials, resource reuse, renewable energy, energy savings, and whole systems thinking.
- The article highlights regenerative software design, hardware design, and data management as potential areas for sustainable innovation.
Table of contents:
Our Earth is being progressively eaten by data
The utilization of Earth's natural resources has always been a driving force behind the progress of our civilization. It's clear that our development depends on them, yet we have never been closer to reaching Earth's limits. As of July 28 of 2023, we are officially drowning in ecological debt. Reckless management of Earth's wealth reflects fundamental lack of understanding ecological boundaries of our planet.
Therefore, it is essential to recognize that resource extraction and digitalization are interlinked in several ways. One positive effect of digital technology, is that it may be used to streamline operations and cut down on unnecessary waste and expenses. On the other hand, virtualization can also lead to increased resource consumption through so-called: rebound effects. This paradox deals with the fact that improvements in efficiency often lead to cost reductions that provide the possibility to buy more of the enhanced product which, in turn, leads to increase in consumption and overuse of natural resources. The best illustration of this paradox is the e-commerce market, where growing demand for online purchases has lead to increased packaging and returns, which resulted in higher energy use per product.
Even though, most of the operations we perform online seems invisible, they are backed by massive IT systems and infrastructure (the cloud), including hyper-scale data centers which provide digital capabilities like storage and cloud computing services. As digital transformation is an ongoing process, such systems are facing a rapid growth in demand for data exchange, processing, and storing. Due to a widespread adoption of AI and machine learning this trend is expected only to grow, prompting the following question:
Is it possible to keep the current development of digital ecosystems in a way that is not harmful to the environment?
IT’s environmental challenges to achieve sustainability
As we perceive the Internet mostly thorough our device's screens, it's easy to overlook the fact that everything that happens online is actually housed in data centers spread across the globe. Those physical spaces, in which companies house and operate their IT hardware infrastructure, play a fundamental role in our society and digital economy. For every data center worker, five more jobs are being supported elsewhere in the economy. In highly digitalized countries, like the Netherlands, they make up about 25% of GDP.
The, seemingly, unstoppable growth of the sector and its negative environmental impact have drawn the public's interest. There are numerous investigations and measurements that attempt to quantify that impact. The most accessible and well-known indicator is the energy consumption level, counted accordingly as greenhouse gas emissions. Globally, it accounts for 4%, but this metric can significantly vary depending on a country's unique conditions and an energy mix. For example, the data center sector in Germany is far above average and consumes as much as 10% of the country’s yearly energy resources.
It is also worth remembering that in this system, data centers and our devices must somehow communicate with each other. When a request is made to transmit data from one device to another, various physical and virtual components respond and handle the query. To provide internet access between continents, underwater submarine cables connecting continents and their countries, mainlands to islands, and islands to each other are required. This infrastructure is considered to have a relatively small ecological footprint. However, impact assessment studies for various cable projects have identified underwater noise and magnetic fields as major issues that have not yet been adequately addressed by the industry.
In order to grasp the environmental challenges of IT systems, it is important to realize that energy consumption is not enough to measure the environmental impact—it’s just the tip of the iceberg. So, how can we reach the deeply hidden underwater glacier shell?
A method called a life cycle assessment (LCA), might prove to be useful in this case. This sustainability strategy is applied to fully understand how any type of product in it's full-life cycle may affect the environment. LCA takes into account all activities from the extraction of necessary raw materials, like metals and minerals, through the manufacture, use, and disposal of it. Let's take a closer look underneath the glacier's surface to see what else contributes to the environmental impact of IT?
Rare earth minerals and planetary boundaries
Most electronic devices, like computer servers (which store data), are typically made of approximately fifty types of materials. These include: metals, thermoplastics, paper, and several types of Critical Raw Minerals. Those resources have high technical, as well as financial value, because of their limited reserves in the earth's crust. They are mostly located in geopolitically sensitive locations with a high prevalence of child labour, and terrifyingly low recycling rates, with only 5% of them retrieved from end-of-life devices.
Rare earth minerals are used not only in servers, they can be found in hard drives, speakers, smartphones and LCD screens just to name a few. Due to the relatively short lifespan and complex material composition of those IT components, their ecological impact contributes to the ever-growing e-waste.
As data centers operate 24/7, they tend to utilize their equipment quicker than regular users, like us, which accelerates the e-waste problem on a global scale. The most common forms of electronic waste generated by digital infrastructures contain racks, computing equipment, displays, circuits, and other electrical components of laptops. Unfortunately, only around one-third of worldwide data center operators use certified recycling partners to manage the waste responsibly.
Rapid technological innovation, shorter product life cycles, and increased consumer demand for electronics only escalates the problem. The volume of e-waste created in wealthy countries is far greater than their capacity to manage it responsibly. Inadequate management infrastructure in Global South nations, where the majority of electronic waste ends up, leads to improper handling of used electronics, which causes irreversible damage to health and the environment.
In order to cool down the heat generated by servers, data centers consume a significant quantity of water, contributing to its' massive and unsustainable consumption. An average data center consumes 11-18 million of liters of water yearly. This is comparable to the water consumption of a city of 30,000–50,000 people.
Uninterrupted access to water is increasingly being recognized as a risk to the digital infrastructure's operation. A decreasing water supply can pose a risk to the continuous access to data and contribute to the depletion of groundwater resources. Considering that some communities regularly face water scarcity due to increasing population, decreasing precipitation, climate change, and poor water management, it is important to keep a close eye on water-greedy data centers, which might become a further driver in amplifying those problems.
Land footprint and biodiversity
Vast quantities of land are also required for the development and IT’s daily operations. As data center companies buy properties all around the world, the advent of digitization has boosted interest, competition, and the value of land. The number of data centers worldwide can be counted in the hundreds of thousands, and infrastructure of this type may house tens of thousands of servers, and use more resources than a whole country.
IT systems affect biodiversity both positively and negatively. Technological advancements have a long history of helping preserve biodiversity through collecting and managing environmental data and developing tools like digital twins. However, it can also have negative impacts through its rapid expansion, which has consequences such as habitat fragmentation and biodiversity loss. In recent years, the data center market has been growing rapidly, fuelled by the developments in AI.
Dark data: digital waste
Dark data refers to data that is collected, processed, and stored but not used for any purpose. It is a growing concern for data centers as it threatens data protection and privacy but also creates redundant hardware demand. Dark data storage is often out of sight and out of mind for many businesses, making it difficult to identify and measure. On average, 52% of the data stored by organizations is 'dark.' Its content and value are unknown, and it is essentially useless until its worth is determined. As the volume of data grows exponentially and the carbon footprint of cloud environments increases, dark data can wind up strewn across numerous cloud service provider regions, making it difficult to comply with security, privacy, and data compliance laws and regulations.
Big Tech hegemony
Big Tech companies have a significant presence in the data center infrastructure industry, which has been growing rapidly in recent years. The concentration of infrastructure in the hands of a few large companies raises concerns about the potential for monopolistic global business practices and a lack of competition, which could lead to increase in prices and decrease in innovation efforts. The dominance of Big Tech companies in the IT systems industry also raises concerns about data privacy and security, as these companies have access to vast amounts of our personal data.
It’s not a secret that AI algorithms, which those Big Tech leaders were secretly developing, have been trained and systematically enhanced using data extracted from society. Almost all of the independent AI research projects, including those purposefully formed as alternatives, have fallen right into the hands of the corporations from whom they were trying to become independent: OpenAI to Microsoft, DeepMind and Anthropic to Google, Hugging Face to Amazon. Many AI researchers view this situation as extremely dangerous.
Regenerative solutions and best practices for sustainable information technology
With such a diverse set of challenges, we must explore eco-friendly solutions and best practices that address technical, environmental and social concerns while additionally bringing up a paradigm shift in how we perceive and form our relationships with digital ecosystems.
And this is where regenerative design comes to play. It is a comprehensive approach to developing greener, efficient ways that aim to reduce the negative impact and actively enhance the resources and natural ecosystems that sustain a prosperous life for all living beings. This philosophy goes beyond conventional sustainable practices and design by aiming to develop and deploy systems that restore or renew their own sources of energy and materials, mimicking nature itself. The key principles of regenerative design include safe and healthy materials, resource reuse, renewable energy, energy savings, and whole systems thinking.
While regenerative design is a broad concept, it generally refers to an approach to development and planning that prioritizes the well-being of people and nature. Let's check how those principles can be applied in a digital ecosystems.
Regenerative software design
Electronic devices have a limited life span and irresponsible development and use of software contribute to their obsolescence. Open Source and Green Coding, which are crucial ingredients in regenerative software and sustainable IT also known as Green IT, benefit the environment and society in multiple ways. This can be achieved by decreasing energy cost and use while increasing the productivity of computer systems. Businesses and developers could easily do their part to reduce the negative effects of digitization by adopting a few simple methods in their daily operations:
- Optimizing the code to consume less power is a crucial first step. For example, we can increase energy efficiency by using fewer loops and fine-tuning the algorithms.
- The choice of programming language can also impact the environmental impact of an application, as some languages are more energy efficient than others.
- Lean coding principles can also be used to minimize the processing power needed to deliver similar results and application, which in turn will minimize e-waste and reduce carbon emissions. More detailed principles on Lean Software Development can be found under this link.
- Additionally, Open source philosophy of software development engages like-minded users and developers in a collaborative sustainability effort to improve the coding practices and share them within the community. This kind of expertise exchange reduces unnecessary duplication while providing cost savings and scalable solutions independent from Big Tech hegemony.
Regenerative hardware design
Due to their unique biological structure biomaterials are considered the most suitable material to realize the emerging generation of environmentally friendly electronic products. Natural materials such as resins, gums, silk, snails, the eggs of quails, and cellulose are becoming more popular for use in electronics. They can be implemented in a range of degradable circuit boards to organic electronics that are infinitely recyclable and reusable.
It is worth mentioning that fungi electronics play a crucial role in research that explores their use to create regenerative electronic devices. These mysterious creatures have unique properties that make them suitable for this purpose, such as their ability to conduct electricity and grow in designated patterns. In the near future we could use them to make biodegradable electronic devices and apply them to various applications, such as sensors, batteries, and even biodegradable circuit boards.
The use of biological materials and reducing the use of hazardous materials in electronics is a growing area of research, and it has the potential to revolutionize the way we think about our devices, their life cycles, and their environmental impacts.
Regenerative data management
An excellent example of a regenerative data management solution that does not endorse big tech hegemony but instead engages and enriches the community's sovereignty is the Wildland Project. It aspires to provide a decentralized, secure, and user-centric solution to data management and sharing independent of any specific service provider or platform.
Overall, the main goal of this project is to provide a new ESG management paradigm that can give us more control over our data while also providing flexible and scalable assets for building innovative apps and services.
This concept comes along with a governance model called UDO (User Defined Organization), which promotes sustainable financing for user-facing projects by enabling the community to participate in decision-making processes related to the development and management of Wildland's infrastructure. Community members and engaged developers can make transparently independent decisions about how their data are stored and managed and what new features are being developed and implemented.
What the future holds for sustainable technology?
For years, IT systems have been constructed with a little thought about their impact on the environment and society. As a result, the foundation of today's digital culture is built on credit borrowed from future generations, in a non-sustainable way, overusing natural resources and creating conflict around this issue. Is there anything we can do, to ensure that the world we will pass to them, will be sustainable?
Communities evolve through communication and information exchange, therefore IT systems will increasingly play a significant role in shaping our society. It is essential to re-evaluate our current relationship with information and communication technologies and steer their advancement in a direction that prioritizes the welfare of people and the planet. Rejecting technologically driven expansion, slowing down the speed of technological change, and adhering to the rhythm of natural cycles could be the first step in making that happen. Only then we will be able to establish a new mindset and fix our reality in such a manner that prioritizes interspecies harmony over profit and convenience.
Lehdonvirta Vili, Oxford Internet Institute: “Cloud Empires: How Digital Platforms Are Overtaking the State and How We Can Regain Control”, The MIT Press, Cambridge, Massachusetts, 2022
Jussila Jari Juhani, Siintoharju Sanna-Maaria: “Regenerative Design in Built Environment”, February 2023
Kunkel Stefanie, Tyfield David: “Digitalisation, sustainable industrialisation and digital rebound – Asking the right questions for a strategic research agenda”, Energy Research & Social Science, Volume 82, December 2021, 102295
Erlandsson Jennifer, Bergmark Pernilla, Höjer Mattias: “Establishing the planetary boundaries framework in the sustainability reporting of ICT companies – A proposal for proxy indicators”, Journal of Environmental Management, Volume 329, 1 March 2023, 117032
Al Kez Dlzar, Foley Aoife M., Laverty David, Furszyfer Del Rio Dylan, Sovacoole Benjamin: “Exploring the sustainability challenges facing digitalization and internet data centers”, Journal of Cleaner Production, Volume 371, 15 October 2022, 133633
Verdecchia Roberto, Lago Patricia, de Vries Carol: “The future of sustainable digital infrastructures: A landscape of solutions, adoption factors, impediments, open problems, and scenarios”, Sustainable Computing: Informatics and Systems, Volume 35, September 2022, 100767
Desislavov Radosvet, Martínez-Plumed Fernando, Hernández-Orallo José: “Trends in AI inference energy consumption: Beyond the performance-vs-parameter laws of deep learning”, Sustainable Computing: Informatics and Systems, Volume 38, April 2023, 100857
Fraunhofer-Gesellschaft, Fraunhofer IZM “Project GreenIT”
Rutkowska Joanna, Zawistowski Julian, and Regulski Andrzej Golem Foundation “Project Wildland: The Why, What, and How”