Research Finder
Find by Keyword
1/26/2026
Key Highlights
The Dawn-Dusk orbit provides nearly 100% solar duty cycles - the sun never really ‘sets’ - thus eliminating the battery weight penalties that plague traditional low-earth orbits.
Terrestrial power constraints are the primary market driver, with global data center capital outlays projected by McKinsey to reach $6.7 trillion by 2030, powering all of that tech is the key limiter.
Orbital solar offers an efficiency advantage over Earth-based power arrays, but cost-per-kilowatt and dropping launch costs won't be enough to make the orbital economic cases alone.
Strategic "slot grabbing" in Sun-Synchronous Orbits (SSO) would deliver first-mover advantages for hyperscalers facing 7-12 year interconnection queues on the ground.
Radiative cooling systems are being architected to utilize the deep-space heat sink, intending to solve the extreme thermal densities of next-generation GPU clusters, even so, cooling is the key constraint compared to ground-based or undersea alternatives.
When you look at the current trajectory of artificial intelligence, it becomes clear that we are architecting a digital infrastructure that the terrestrial grid was never designed to support. The persistent hunger for power has pushed humanity closer to the edges of our geographic and social limits. Our analysis suggests that the next logical step is not just a change in location but a change in altitude. In orbit, the Dawn-Dusk orbit is the type of ambitious answer to the mounting environmental and regulatory pressures facing the AI industry.
In a Dawn-Dusk configuration, a satellite gets to effectively "ride the terminator," the dividing line between day and night. This specific 98 degree Sun-Synchronous Orbit allows a data center complex to remain in constant sunlight. On Earth, data center solar is at the mercy of the day-night cycle and weather patterns. In this specific orbit, solar arrays have the capability to be "always on." This eliminates the need for massive chemical battery arrays, which often account for a significant portion of a satellite's mass. By shedding this weight, operators can instead fill the launch fairing with high-density GPU clusters.
The engineering shift is fundamental. While terrestrial data centers spend millions of gallons of water on cooling, an orbital facility would use passive radiative cooling. One side of the craft is designed to absorb solar energy, while the opposite side faces the -270 degree Celsius void of space. This creates a natural thermal gradient architected to facilitate heat dissipation without the mechanical failure points of terrestrial chillers. That said, the size of those cooling arrays create a series of substantial challenges.
Market Dynamics: The $1.4 Trillion Pivot
The shift toward the "Space Edge" is being fueled by a massive redirection of capital. According to recent findings from ARK, investment in data center systems has surged to an annualized growth of 29%. By 2030, this global investment is expected to reach $1.4 trillion. However, the terrestrial "power wall" is making it increasingly difficult to deploy this capital effectively.
Based on our analysis of the market, the Dawn-Dusk orbit becomes more than just a scientific curiosity; it has the potential to become a strategic asset class. In the next few years, we could see a multi-national "gold rush" for Sun-Synchronous slots that mirrors the early days of terrestrial spectrum auctions. The total addressable market for space-cloud computing is valued at approximately $6.12 billion in 2025, but it is expected to hit nearly $25 billion by 2035. This growth could serve two distinct needs: processing the massive volumes of Earth-observation data in situ, and providing an "off-grid" sanctuary for high-value AI training.
The competitive landscape is bifurcating between established giants and agile pure-play startups. Google’s Project Suncatcher and SpaceX’s internet provision service Starlink V3 can theoretically provide the infrastructure backbone, while firms like Starcloud are already operating NVIDIA hardware in orbit to demonstrate the commercial viability of the model. Meanwhile, European initiatives like ASCEND aim to deliver a sovereign, zero-emission compute alternative that bypasses the land-use conflicts; like those currently stalling projects in Northern Virginia and Germany.
Looking Ahead
Based on what we are observing, the Dawn-Dusk orbit has the potential to become the most contested piece of property in the digital age. The key trend that we are going to be tracking is the vertical integration of launch services and data center operations. Where a company owns the rocket, they own the gateway to the prime off-world locations. Right now, that favors SpaceX and (soon) AWS/Blue Origin. Analysis suggests that as launch costs continue to fall with widespread use of large re-usable rockets, orbital compute comes closer to a solid business case compared to restricted terrestrial facilities.
Based on analysis of the market, our perspective is that the "terminator economy" will eventually lead to a decoupled compute model. In the future, the primary "brain" of a frontier AI model may live in constant sunlight at 20-50 millisecond latency, while the "eyes" and "ears" remain on the ground to reduce that latency for inferential AI. Going forward we will be tracking how companies perform on the development of in-orbit robotic assembly, a necessary step to scale these facilities beyond the size of a single launch fairing. We will also be looking at alternatives, such as undersea data centers, to reduce the impact of data centers on land. Microsoft tested this with Project Natick, which the company ended in 2024 to focus on land-based data centers. Advances in tidal energy and cooling in the ocean will provide a solid alternative to both land and space.
This is all medium-future speculation, but when you look at the regulatory environment tightening on the ground, movement toward space is a direct response to the perceptions of a finite Earth. HyperFRAME Research will be tracking how the leading hyperscalers balance their terrestrial expansion with these high-risk orbital investments in future quarters. The question is no longer whether we can build data centers in space, the technology exists, now it will be who can do so in a cost-effective way.
Stephen Sopko | Analyst-in-Residence – Semiconductors & Deep Tech
Stephen Sopko is an Analyst-in-Residence specializing in semiconductors and the deep technologies powering today’s innovation ecosystem. With decades of executive experience spanning Fortune 100, government, and startups, he provides actionable insights by connecting market trends and cutting-edge technologies to business outcomes.
Stephen’s expertise in analyzing the entire buyer’s journey, from technology acquisition to implementation, was refined during his tenure as co-founder and COO of Palisade Compliance, where he helped Fortune 500 clients optimize technology investments. His ability to identify opportunities at the intersection of semiconductors, emerging technologies, and enterprise needs makes him a sought-after advisor to stakeholders navigating complex decisions.