One of my responsibilities while working at the Defense Intelligence Agency (DIA) was to research and write papers on potential disruptive technologies and their impact on the Intelligence Community. It was in this role that I first encountered the concepts of DNA data storage and quantum computing. Fast forward to today: I now represent HD-28 in the Virginia General Assembly, home to at least 112 data centers, and where 70% of the world’s internet traffic flows every day.
While data centers have brought high-paying union and non-union jobs, economic development, and real estate tax reductions to Loudoun County, my constituents still routinely ask: “Why can’t we just put all these data centers in orbit?”
Drawing upon my experience at DIA, the National Reconnaissance Office – which launches the nation’s spy satellites, and as the former global product manager for AT&T’s high-speed global network, it’s important to recognize that placing data centers in space would require transformational advances in a wide range of technologies – most notably in data storage, network connectivity, and quantum computing.
Currently, the modern interconnected global economy relies heavily on silicon-based memory storage, fiber-optic cables, and operating systems designed to function in climate-controlled environments and shielded from harmful space radiation by the Earth’s atmosphere and magnetic field. Let’s look at some of the challenges that would need to be overcome to put data centers in space.
Accessing Space is Hard
Only five U.S. states have vertical launch capabilities: Virginia, Florida, Texas, California, and Alaska. Even today’s most powerful, reusable commercial rockets cost tens of millions of dollars for each launch.
To place data centers in orbit will require an increased tempo in launches, at dramatically reduced costs. While companies like SpaceX and Rocket Lab are pushing the boundaries of frequency, reliability, and reusability, the routine deployment of data center-sized infrastructure into low earth orbit remains decades away.
Space is Dangerous
In J.J. Abrams’ Star Trek, Dr. Leonard “Bones” McCoy characterizes the dangers of space: “Space is disease and danger wrapped in darkness and silence. One tiny crack in the hull and our blood boils in thirteen seconds. Solar flare might crop up, cook us in our seats. Space is dangerous.”
This danger isn’t just limited to people – it applies equally to the sensitive electronics necessary to run today’s data centers. By some estimates, a single radiation hardened microprocessor could cost upwards of $200,000, instead of $300 for an Earth-based processor.
None of the current data center technology is designed to operate outside the protection of Earth’s atmosphere. Therefore, we can’t simply take a server rack, launch it into space, and expect it to work. Just one server rack might cost up to $750,000 to launch and in an average sized data center, there might be 2,500 such server racks.
Without radiation hardening, and a significant increase in cost, they would quickly become just more space junk.
Rethinking Storage and Computing
Future technologies might make it feasible – might – to use fewer launches, but regardless, the technology will still need to be hardened to address the deadly environment in space. For instance, DNA storage theoretically could store all of humanity’s knowledge in a box no larger than a home refrigerator.
The problem is that encoding, writing, and reading data from DNA storage is extremely slow. Quantum computing could dramatically accelerate data processing and the read-write processes of DNA storage, but the operational use of these technologies is still decades away.
Space-Based Networking Challenges
Even if we solved the storage and computing challenges, we’d need dramatically improved connectivity between orbit and the ground. Current networking technology (5G) isn’t designed for space. New networking standards (7G), using laser-based communication and ultra-high-frequency signals, would have to be developed and deployed.
Although there’s active research underway, these advancements are also years from operational, reliable, widespread availability.
Collectively, the convergence of these groundbreaking technologies – quantum computing, DNA storage, and ultra-high-speed orbital networking – may be 75 to 100 years from integrated, practical implementation.
But as President Kennedy reminded us: “We choose to go to the Moon… not because it is easy, but because it is hard.” The same principle applies here. Moving data centers to low earth orbit would be complex, but that shouldn’t discourage us from beginning the discussions now.
Our future in space, like our future on Earth, belongs to those who dare to ask the hardest questions – and who are willing to invest the time, creativity, and resources to find the answers. I’m proud to represent a district that not only powers today’s internet but also has constituents who are beginning to articulate a vision for tomorrow’s infrastructure.
