Even though the technology that powers them has evolved in terms of performance and efficiency, the fundamental principle of data centers hasn’t changed much over the past two decades.
Each new generation of infrastructure has been designed to shorten the distance to users, thereby reducing latency and increasing reliability. However, AI is seriously challenging this model, and unprecedented demand for data centers is prompting businesses to think outside the box.
Land-based projects now face challenges from all sides, including land availability, cooling, electricity and water consumption, and local opposition.
Enter the orbital data centers. Some pioneers are now considering projecting data centers into space, and the model continues to move further from concept and closer to reality.
Although huge technical hurdles remain, advocates argue that orbital data centers could eventually complement terrestrial campuses by moving some of the most power-hungry AI workloads away from terrestrial constraints.
Networking is what will define a good orbital calculation
Some of the benefits include virtually unlimited space to spread out, uninterrupted access to green solar energy, and cooler environments that don’t require as much cooling.
But while sending calculations into space solves some of the problems we face today, it undoes one of the biggest fixes we’ve spent years improving: latency.
In today’s instant economy, a data center is useless if it cannot exchange information quickly and reliably. Terrestrial networks rely heavily on fiber optic networks and physical connections, but orbital data centers are expected to improve wireless transmission over hundreds of kilometers.
As the work progresses, Ivo Ivanov, CEO of internet exchange operator DE-CIX, says orbital computing should be seen as an additional layer in our increasingly distributed digital ecosystem – not an outright replacement for terrestrial computing.
To better understand whether orbital data centers are just another ambitious experiment or whether they could be the next evolution of digital infrastructure, I spoke with Ivanov about the industry’s biggest challenges.
- You noted that the SpaceX IPO brought the idea of AI data centers in space into the mainstream. Leaving the hype aside, is this close to reality and what is really driving the world’s biggest companies to look to space?
The idea of placing data centers in orbit is the kind of thing that immediately captures people’s imagination. It belongs to the same category of technological ambition as reusable rockets, commercial spaceflight or the moon landing.
For decades, it seemed like something that belonged in science fiction. Today, for the first time, it is being discussed as a serious infrastructure proposal. But I don’t think the real story here is about space, but rather about AI and how it is changing our global trajectory.
Over the past few years, we have seen an extraordinary increase in demand for computing capacity. AI training hubs are expanding, energy needs are increasing, and in many regions the availability of energy, land and cooling has become a real constraint.
These constraints drive innovation, and that’s what we’re seeing here.
In theory, space provides continuous access to solar energy and allows us to evolve without the physical limitations we experience here on Earth. That said, we are still at a very early stage.
I don’t think we’re talking about replacing land infrastructure in the near future. What we are seeing today is the start of a serious exploration of how orbital resources could complement the already existing digital infrastructure ecosystem.
There’s still work to be done, especially when it comes to networking: putting an AI data center into orbit would be a big step, but it won’t matter much if we have to carry hard drives into space for training. For now, all we can say with certainty is that this is now part of a broader debate about how the company will support the next generation of AI workloads.
- What problems could orbital data centers potentially solve that terrestrial infrastructure faces today?
I think the broader answer is that orbital infrastructure has the potential to alleviate some of the physical constraints that have shaped and limited digital infrastructure for decades. We can’t just keep building more data centers: they need access to land, power, cooling, transport links and, of course, connectivity.
As AI continues to grow, some of these assets will become more difficult to secure, especially at the pace the industry wants to evolve.
This is why the industry is starting to look for non-terrestrial alternatives. Suddenly you’re talking about access to potentially unlimited solar power, lots of space, and the ability to scale at a pace that matches our AI ambitions.
But putting data centers into orbit, as impressive as that may be, is only the first step. Google’s Suncatcher project is already exploring how solar energy could be harvested, while the European Space Agency’s OFELIAS project is exploring how optical power links between Earth and orbit can be optimized to reduce latency and disruption. There are a lot of moving parts.
- From a networking perspective, what challenges will emerge once data centers are put into orbit?
As soon as you put a data center into orbit, networking becomes a deciding factor. A data center on Earth can typically rely on a dense ecosystem of fiber optic routes, Internet exchanges, cloud on-ramps, and interconnection facilities. In orbit, this ecosystem does not yet exist.
Every workload, every application, and every AI model depends on moving data between different locations, and those journeys suddenly become much more complex.
A lot of the discussion I hear is about bandwidth or speed, but I think predictability is the more interesting challenge. AI systems need data to arrive quickly, but they also need it to arrive consistently.
Optical and laser communications have enormous potential, but they also introduce new variables such as cloud cover, atmospheric turbulence, satellite transfers and changes in orbital positions which can all affect how data moves between Earth and space.
This is why I think the real challenge is not just connecting an orbital data center to the ground, but creating an interconnection layer that allows orbital, terrestrial, cloud and edge infrastructure to behave as if they are part of the same ecosystem.
This is a much bigger networking challenge than people think.
- Elon Musk said orbital infrastructure was “only milliseconds away.” How important can those milliseconds be, and is latency the biggest challenge?
Those milliseconds absolutely matter. AI systems are fundamentally data-driven, and every additional moment spent moving information between users, applications, models, and infrastructure affects performance and productivity.
This is one reason the industry is investing so much in edge computing: the closer you can bring computing resources to the data, applications, and users they serve, the better the experience.
Granted, low Earth orbit is only milliseconds away, but for latency-sensitive use cases, the 20-40ms needed to reach the stratosphere will be a significant challenge for some AI inference use cases. Not all applications can therefore be operated hundreds of kilometers from the ground.
That said, performance isn’t just about latency. A network that consistently delivers twenty-five milliseconds of latency is often more valuable for many enterprise use cases than one that fluctuates between twenty and one hundred.
Physics determines how quickly data can travel, but factors such as atmospheric conditions, satellite handoffs, routing decisions, and network architecture all influence the predictability of that experience.
- If a data center is in orbit and the AI workloads it serves are on Earth, what actually needs to happen for these two worlds to behave as one network?
The way I like to think about it is that users shouldn’t have to think about where a workload is running. Whether an application is served from a terrestrial data center, an edge location, a cloud region, or one day, an orbital platform, the experience must be seamless.
Achieving this is much more difficult than it seems, because data must flow continuously between different environments, and these environments must behave as if they are part of the same network, even when separated by hundreds or thousands of kilometers and, increasingly, the boundary between Earth and space.
This is one of the reasons why projects like ESA’s OFELIAS initiative, which I mentioned earlier, are so important. As part of this initiative, DE-CIX is working alongside partners including the German Aerospace Center to explore how optical feeder links between satellites and the ground can become more stable, efficient and predictable.
Rather than simply creating more connections, we need to create an infrastructure that allows terrestrial, satellite and orbital resources to work together within a single ecosystem.
We have spent decades building this type of seamless interconnection on Earth, and now we must take what we have learned and combine it with new technologies to integrate non-terrestrial infrastructure.
- Where are terrestrial data centers and interconnection? Will orbital calculation compete with what exists on the ground, or depend on it?
I don’t see it as a competition at all. Digital infrastructure has a long tradition of adding new layers rather than r location of existing ones. Cloud computing has not eliminated enterprise data centers, edge computing has not eliminated centralized clouds, and satellite connectivity has not replaced terrestrial networks.
Instead, each technology tends to find the role that suits it best, and the overall ecosystem becomes more efficient as a result.
I think orbital computing will follow a similar path. Some AI workloads could benefit from processing closer to abundant energy sources in orbit, while others will remain in edge terrestrial data centers because they require ultra-low latency inference, regulatory compliance, or proximity to users and applications.
It’s not about choosing one over the other, but about creating an architecture where workloads can be placed where they make the most sense and where data can flow seamlessly between them.
- What should the industry do now to prepare networks for orbital computing?
I think the first step is to recognize that orbital computing is more than a computational challenge. It’s all about networking. The industry has spent decades building highly interconnected ecosystems on Earth, where data can flow efficiently between businesses, cloud providers, networks, content platforms and users.
If computing begins to grow in orbit, we will need to ensure that the same level of interconnection exists between terrestrial, satellite and orbital environments. Advances in satellite communications, laser networks and distributed infrastructure are creating new possibilities that did not exist a decade ago.
What I would like to see now is greater collaboration between network operators, cloud providers, infrastructure companies and space organizations.
The technologies themselves are advancing rapidly, but the challenge is ensuring they evolve as part of a cohesive ecosystem rather than a collection of isolated systems.
- In ten years, what will a fully interconnected ecosystem spanning terrestrial, edge, cloud, satellite and orbital infrastructure actually look like? What technologies will make it work?
The most interesting thing about the future is that people probably won’t think about infrastructure at all. Today, we spend a lot of time determining whether something works in a cloud region, in an edge location, in a data center, or on a particular network.
In ten years, if we succeed with digital infrastructure, workloads will simply be moved to where they can be processed most efficiently, and data will flow across terrestrial, satellite and orbital environments as naturally as it moves across continents today.
When we look back at the great infrastructure achievements of the past, whether railways, power grids, undersea cables, or the Internet itself, their success is measured not by the technology, but by the way in which they have completely disappeared from everyday life.
People stopped thinking about infrastructure and started focusing on what it enabled. I think we’ll see something similar here. The future of digital infrastructure will be defined by our ability to connect every layer of the ecosystem into something that appears seamless.
That’s when we’ll know we’ve succeeded.
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds.
