The recent explosion in the use of Artificial Intelligence (AI) and Generative AI (GenAI) applications has expanded the list of development and operational challenges that accompany modern information systems. The continuous and optimized use of available computing infrastructure, where applications that process large amounts of data can run, has become critical for businesses today.
The Rise of Renting Computing Resources
Cloud computing, meaning the “rental” of computing resources (such as processing power, data storage, and software) over the internet, initially emerged as an infrastructure solution and began gaining widespread recognition after 2006, when Amazon launched its first cloud services, known as Amazon Web Services. In 2010 and 2011, Microsoft Azure and Google Cloud made their respective debuts, marking the establishment of software, platform, and infrastructure-as-a-service offerings for companies of any size. Today, this market continues to grow with new providers, and its value is projected to surge from around $6 billion in 2008 to nearly $905 billion in 2026.
Cloud computing gave both small and large organizations access to theoretically unlimited computing power. At the same time, it acted as a catalyst for the rapid growth of startups, largely due to the pay-as-you-go model, which significantly reduced the upfront capital expenditure required to operate information systems. In other words, instead of purchasing expensive computing infrastructure, a business can simply rent it for as long as it needs it.
More Than 20 Data Centers in Greece
Today, we are witnessing a race to build new data centers across various parts of the world, including Greece. More than 20 data centers are currently operating or under construction, primarily in Attica, Macedonia, and Crete, giving Greek businesses, public sector organizations, and research centers access to powerful computing infrastructure located within Greek territory. This is particularly significant because the closer an information system operates to its data sources and end users, the lower the processing latency. In addition, keeping sensitive data within specific geographic boundaries is equally desirable for privacy and security reasons — for example, in the case of electronic health records.
However, building data centers is heavily dependent on energy infrastructure, energy costs, water resources, and land availability. If these constraints are not addressed strategically, they can pose serious sustainability risks for the entire endeavor. Data centers can occupy an area larger than 100 soccer fields and may consume nearly as much energy in a single year as an entire large municipality. Cooling accounts for 40% of the total operating cost of a data center. This has driven the search for even more advanced solutions, reaching all the way into outer space.
Looking to Space for Solutions
In November 2025, an American startup called Starcloud, in collaboration with Nvidia, launched the first components of an experimental data center into orbit, featuring chips specifically designed for AI applications. The data center will draw its required power directly from the Sun, using panels that orbit alongside the rest of the infrastructure, effectively addressing the physical constraints that exist on Earth.
These developments, along with unexpected events such as a recent drone attack on a data center in the Middle East, highlight the need to make full use of all available computing infrastructure, wherever it may be located, including at the very edges of the network.
The computing continuum (Cloud-Edge Continuum) is being used more and more to host big data and AI applications, leveraging multiple clouds as well as highly heterogeneous devices located at the network’s edge, such as infrastructure tied to 5G antennas, autonomous vehicles, and drones.
Greece’s Contribution
In a recent research effort, NebulOuS, a European research and development project involving Greek researchers, focuses on this scientific domain. The initial concept was developed by scientists at the Research University Institute of Communication and Computer Systems (ICCS) and the Athens University of Economics and Business. NebulOuS was designed as a middleware information system that simplifies the interaction between AI and IoT (Internet of Things) applications — software and systems that connect physical devices to the internet, allowing them to collect and exchange data and automate operations — and the heterogeneous infrastructure resources of the computing continuum. It is an innovative meta-operating system that seamlessly exploits edge devices in combination with the abundant cloud computing resources available from multiple providers simultaneously. A core concept here is the “brokerage” of computing continuum services (Cloud, Fog, and Edge), which facilitates the reduction of the variability space when searching for the optimal application topology at any given time, taking into account users’ quantitative and qualitative preferences. What makes this particularly interesting is that, through this brokerage mechanism, the platform allows small and medium-sized enterprises to offer their available edge infrastructure for “rental” and to become part of third-party application topologies in a secure and reliable way.
Multiple Applications for NebulOuS
The innovations of NebulOuS have already been tested and validated across a range of application domains, including smart city applications, precision agriculture, crisis management, IoT-based energy management, predictive analytics for urban sustainability, and more. NebulOuS was recently completed successfully, with the participation of 16 organizations from across Europe, including universities, research centers, and private companies.
What is certain is that innovations like these demonstrate that the computing continuum can and must be effectively utilized, offering significant advantages over current information systems that European businesses are called upon to develop and operate.
Giannis Verginiadis is an Assistant Professor in the Department of Management Science and Technology at the Athens University of Economics and Business, a researcher at the Research University Institute of Communication and Computer Systems (ICCS), and the scientific lead of NebulOuS.
