Orbital Chenguang’s Plan to Build China’s Orbital Data Centers and Reach Gigawatt-Scale Computing by 2035
Orbital Chenguang aims to develop space-based data centers with 57.7 billion yuan in credit lines, targeting a gigawatt-scale orbital network by 2035.
China’s Next Data Center Boom Might Orbit the Planet
Orbital Chenguang, a Beijing-based startup, has emerged as a central actor in an ambitious Chinese effort to move some computing off Earth and into low Earth orbit. Backed by early-stage funding and strategic credit lines totaling 57.7 billion yuan (roughly $8.4 billion) from a dozen major Chinese banks, the company is positioning orbital data centers as infrastructure intended to relieve the mounting constraints facing terrestrial data centers — from energy and cooling to land use — as demand for cloud services and artificial intelligence grows.
The proposal is being pursued inside a state-linked ecosystem: the Beijing Astro-future Institute of Space Technology coordinates a consortium of 24 companies and research groups supporting the effort. That alignment of government, finance, and private actors marks the project as more than a fringe experiment and signals how seriously China is treating space-based computing as a potential national priority.
Why Move Data Centers into Space
Orbital Chenguang’s pitch responds directly to limitations afflicting Earth-bound computing. Modern data centers consume large quantities of power, require complex and costly cooling systems, and occupy valuable land — pressures that intensify as AI workloads and cloud demand scale. By placing computing assets in a Sun-synchronous orbit roughly 700 to 800 kilometers above the planet, the initiative aims to take advantage of near-continuous solar exposure for power and the ambient thermal conditions of space to potentially ease some cooling burdens.
The concept is not simply about novelty; it’s framed as an infrastructural pivot. A distributed network of computing satellites could, in theory, handle some classes of workloads while reducing the need to expand terrestrial footprint and grid capacity. Orbital Chenguang’s stated long-term ambition is to assemble a gigawatt-scale computing system in orbit by 2035 — a goal that would require dramatically larger solar arrays and a coordinated constellation of satellites.
Financial and Institutional Backing
The project’s financial architecture is anchored by the 57.7 billion yuan in strategic credit lines from multiple Chinese banks. While those credit lines are not the same as immediately available cash, they represent a significant vote of confidence and signal priority at both financial and institutional levels. The Beijing Astro-future Institute of Space Technology — a government-backed organization — is coordinating a 24-member consortium of companies and research groups, embedding Orbital Chenguang within a broader national effort that pairs public policy with private innovation.
This state-linked ecosystem mirrors a pattern in which long-term, high-capital projects tied to space and advanced computing benefit from alignment between government, industry, and finance. That alignment can ease access to capital, facilitate regulatory coordination, and integrate the project into wider national strategic planning.
Technical Approach: Orbit, Power, and Thermal Strategies
Orbital Chenguang intends to operate satellites in a Sun-synchronous orbit at approximately 700–800 kilometers altitude. That orbital choice provides near-continuous exposure to sunlight across parts of each orbit, which the project plans to exploit for sustained solar power generation. The thermal environment of space also factors into their rationale: without atmospheric convection, radiative cooling becomes the sole mechanism for heat rejection, and the company’s materials and thermal-design strategies would need to address that fundamental engineering constraint.
The envisioned pathway to a gigawatt-scale system implies solar arrays and power-management systems orders of magnitude larger than current space deployments. Achieving sustained, large-scale power generation in orbit would also require solutions for array deployment, power distribution across a constellation, and long-term survivability in a high-radiation environment.
A Phased Roadmap Toward 2035
Orbital Chenguang’s plans are framed as a multi-stage rollout spanning the next decade and beyond. The near-term phase, running from 2025 to 2027, focuses on core engineering challenges and early satellite testing. This stage is intended to validate subsystems and test the feasibility of space-based computing under real orbital conditions.
From 2028 to 2030 the roadmap calls for integrating space-based computing with terrestrial networks — building a hybrid infrastructure that distributes workloads between Earth and orbit. If these phases succeed, the plan targets a fully scaled, gigawatt-level constellation by 2035. The company’s planned Chenguang-1 test satellite, however, has not yet been confirmed for launch, underscoring that demonstrations and actual deployments remain forthcoming.
Other Players and an Emerging Space Edge Ecosystem
Orbital Chenguang is not operating in isolation. Other Chinese entities are already pursuing space-based edge computing. The report notes that ADA Space and Zhejiang Lab launched a 12-satellite edge computing constellation in 2025, and that several startups are developing smaller-scale computing satellites and demonstration missions. Those efforts are laying technical groundwork and building a nascent ecosystem for compute-at-the-edge in space.
China is also investing in the launch infrastructure and vehicle capability needed to support large constellations: reusable rocket development, new spaceport construction, and filings with the International Telecommunication Union that indicate intentions for large-scale networks. Together, these developments suggest a coordinated buildup of launch, in-orbit, and ground segments necessary to support ambitious orbital computing visions.
Engineering and Operational Challenges
The technical hurdles are substantial and acknowledged in the project’s public descriptions. Thermal management is a central challenge: in vacuum, heat must be carried away by radiation rather than convection, which requires large and highly efficient radiator surfaces. That constraint complicates cooling designs for high-density computing hardware.
Power generation and deployment present another set of obstacles. The solar arrays and energy-storage systems needed to approach gigawatt-scale operations would be vast and complex to launch and assemble. In-orbit deployment and maintenance of such structures would require advances in assembly methods, long-life power electronics, and possibly new logistics for modular upgrades.
Radiation exposure in low Earth orbit also threatens sensitive electronics. High-energy particles can cause bit flips, degrade components over time, and require resilient system architectures, radiation-hardened hardware, shielding strategies, and fault-tolerant software systems.
Each of these challenges — thermal control, power scaling, and radiation resilience — multiplies when moving from demonstration satellites to a large, coordinated constellation intended to support production workloads.
How the Proposed System Would Work in Practice
From the descriptions provided, the intended operational model is a hybrid infrastructure in which space-based computing coexists and cooperates with terrestrial data centers. Early testing will focus on engineering validation; subsequent phases will attempt to integrate orbital assets with Earth-based networks so that workloads can be distributed depending on latency, power availability, cooling needs, or other constraints.
Near-continuous solar exposure in Sun-synchronous orbits is a core part of the power strategy, while radiative thermal management is positioned as the primary cooling mechanism. The project’s long-term scaling envisions thousands or more satellites working together, with solar arrays and radiators sized to support increasingly heavy compute loads.
Who Stands to Use Orbital Computing and Why It Matters
The source frames orbital data centers as infrastructure intended to relieve pressures on power, cooling, and land use driven by surging AI and cloud demand. Organizations constrained by terrestrial energy availability or land-use limitations — or those requiring distributed or resilient compute nodes — are implied beneficiaries of a hybrid approach that can place select workloads in orbit.
More broadly, the development may matter to cloud providers, enterprise IT planners, researchers in AI and high-performance computing, and national infrastructure planners who are watching how compute demand shapes energy and land use. The project’s integration into a state-backed ecosystem suggests potential interest from government and large institutional users alongside private firms.
Implications for Developers, Businesses, and the Industry
If Orbital Chenguang’s plan advances beyond testing, the broader industry implications could be significant. A viable orbital compute layer would shift considerations for capacity planning, edge-compute strategies, and long-term infrastructure investment. Developers and system architects would need to adapt applications and deployment models to exploit distributed orbital compute nodes, and businesses would need to assess cost, latency, regulatory, and operational trade-offs compared with terrestrial cloud and edge solutions.
At the same time, the emergence of large orbital compute constellations would intersect with related ecosystems: satellite communications, launch providers, reusable rockets, spaceports, and regulatory processes such as ITU filings. Security, data sovereignty, and compliance frameworks would also become central discussion points for enterprises considering the use of space-based compute resources.
Broader Strategic and Industry Context
The undertaking is being treated as a strategic priority inside China’s broader space ambitions. The involvement of a government-backed institute coordinating a consortium, the substantial credit-line support, and national investments in launch capability all situate the project inside a coordinated push to expand China’s in-orbit infrastructure.
This effort also dovetails with parallel technology trends: growth in AI workloads that increase demand for power and cooling; interest in edge computing to reduce latency and distribute processing; and expanded launch capacity that makes larger constellations more feasible. Together, these factors form a permissive environment for experimentation with nontraditional approaches to data-center design and siting.
Practical Uncertainties and Known Unknowns
Publicly available details leave several important questions open. The credit lines secured by Orbital Chenguang are strategic but not the same as immediately deployable capital. The Chenguang-1 test satellite has not been confirmed for launch, indicating that in-orbit technical validation remains a forthcoming milestone rather than a completed one. The timeline toward a gigawatt-scale orbital system is aspirational, and the plan depends on solving multiple hard engineering problems while scaling launch and in-orbit servicing capabilities.
How This Fits with Related Technologies and Ecosystems
The orbital data center concept relates to other technology domains that the industry already tracks: AI hardware platforms designed for edge and harsh environments, satellite communications for data transport, launch-vehicle innovations, and space systems for power and thermal control. The source links the idea to contemporaneous industry work, including other Chinese constellations and early demonstrations of computing in orbit. Those adjacent developments form potential paths for collaboration, competition, and standards-setting as the idea matures.
Regulatory and Operational Considerations
Although detailed regulatory plans are not spelled out, the mention of International Telecommunication Union filings indicates preparatory steps toward spectrum and orbital coordination for large networks. Operating a sizable orbital data constellation would require navigation of frequency allocation, space traffic coordination, and national and international rules that govern satellite operations and data transfer.
Industry-Level Impact and Potential Scenarios
If Orbital Chenguang and the broader Chinese ecosystem successfully demonstrate reliable space-based computing modules and integrate them into Earth-based networks, multiple scenarios could unfold. One possibility is that orbital compute serves niche, high-value workloads that benefit from distributed nodes or specific environmental conditions. Another is a more transformative shift in which orbital resources relieve peak terrestrial demand and change long-term planning for power and cooling in cloud architecture. Yet the technical and logistical challenges mean that each scenario rests on substantial, demonstrable progress in engineering, launch cadence, and in-orbit operations.
China’s approach — pairing state-backed coordination with startup innovation and bank credit lines — also offers a model that could accelerate progress relative to purely private-sector efforts, at least in terms of resource mobilization and national-scale planning.
The broader market reaction and international responses will depend on how quickly test missions move into routine operations and whether other countries and companies pursue comparable orbital compute experiments.
Looking ahead, the pathway from concept to operational orbital data center remains long and technically demanding. Demonstrations such as ADA Space and Zhejiang Lab’s 2025 12-satellite edge constellation and the planned tests by Orbital Chenguang will be important early indicators of feasibility. Continued investments in reusable rockets, launch capacity, and spaceport infrastructure will factor heavily in determining whether the aspirational timeline to hybrid Earth-orbit systems by 2035 can be realized. The coming years of engineering tests, regulatory filings, and incremental deployments will determine whether computing in orbit becomes an experimental curiosity or a genuine extension of global computing infrastructure.
