If you were to move a data center into space, how many steps would it take?

Step 1: Send the computers into space. Step 2: Enable unified software management of these computers. Step 3: Connect them via a high-speed network and transmit data back to Earth.

The two satellites recently launched into orbit by the Lijian-1 Yao-18 mission are doing just that.

The Chaozhisu-1 satellite carries a 4-meter visible light camera and an AI computer, enabling on-orbit image processing and target identification. The Pengcheng-NSK-1 satellite integrates 5G NTN, on-orbit computing, and satellite-ground laser communication into a single satellite, with plans to validate a 100 Gbps laser link in the future.

One satellite addresses 'computing,' while the other integrates 'computing' and 'connectivity.'

The history of computer development has repeatedly proven one thing: stand-alone performance sets the upper limit, while the network determines scalability. From mainframes and local area networks to the Internet and cloud computing, computing power has truly become infrastructure not just because processors have gotten faster, but because machines can connect with each other, tasks can be scheduled, and data can flow efficiently.

Now, this logic is entering orbit.

China is incrementally sending into space the capabilities required for a data center.

01

The first piece, of course, is the computer.

In May 2025, the Zhejiang Lab-led Santih Computing Constellation launched its first 12 computing satellites into orbit, with each satellite achieving a peak computing power of 744 TOPS and the initial mission delivering a total on-orbit computing power of 5 POPS. By February of this year, the team had established a network among six satellites in orbit and achieved unified management and scheduling of computing, storage, and network resources through a space-based distributed operating system.

More important than 5 POPS is that computing resources are beginning to transcend single-satellite boundaries.

The Santih Computing Constellation has deployed multiple AI models and applications. One model for astronomical observations can directly judge and classify gamma-ray bursts on orbit, reducing the daily data transmission requirements from hundreds of MB to tens of KB and shortening processing time from hours to seconds.

This is precisely why edge computing exists. Satellites generate vast amounts of data daily, but many operations do not require transmitting all raw data back to Earth for processing. If sufficient computing power is available in orbit, data can first be filtered, identified, and compressed, with only the most valuable information sent down.

However, as the number of computing nodes increases, another question arises: How do you install, update, and adapt software on these machines for different tasks?

02

The Tiansuan Constellation, driven by Beijing University of Posts and Telecommunications, has shifted its focus to software and resource management.

The BUPT-1 satellite has brought cloud-native technology to satellites, conducting experiments in distributed on-board AI inference and on-board 5G core networks. The BUPT-2 satellite has launched a self-developed high-performance server as a computing payload into orbit, validating capabilities such as dynamic updates of on-board containers and incremental model updates.

Such work, while far less eye-catching than 'hundreds of TOPS,' determines whether space-based computing power can be used at scale.

Traditional satellites are highly customized, with software, payloads, and hardware tightly bound, and missions typically determined before launch. Computing infrastructure requires a different capability: software can be updated, applications can be redeployed, different hardware can be uniformly managed, and computing resources can be scheduled according to tasks.

In today's data centers, no one redevelops software for the entire business just to add a server. If space-based computing power cannot achieve this in the long run, each computing satellite will remain an expensive specialized device, with management costs rising as scale increases.

Thus, after TOPS, operating systems, software environments, and resource scheduling have taken center stage. This is not a conceptual upgrade but foundational work essential for any large-scale computing system.

03

Once computing resources can be managed, bandwidth quickly becomes the next bottleneck.

In January of this year, the Aerospace Information Research Institute of the Chinese Academy of Sciences used AIRSAT-02 to conduct a ultra-100G satellite-ground laser communication experiment, increasing the communication rate from 60 Gbps to 120 Gbps through on-orbit software reconfiguration without changing satellite hardware. The maximum continuous communication time reached 108 seconds, during which 12.656 Tb of data was transmitted.

As computing power increases, the pressure on data transmission becomes apparent sooner.

If processor performance continues to rise while communication capabilities stagnate, data will still accumulate in orbit. Ground-based AI data centers have already demonstrated this issue in recent years: after the explosive growth of GPU performance, high-speed interconnects rapidly shifted from a supporting role to a core capability, as thousands of chips could only work efficiently as a cluster if data exchange was fast enough.

Conditions in space are even harsher. Satellites move at high speeds, satellite-ground visibility is limited, and laser communication is affected by weather, atmospheric conditions, and acquisition and tracking accuracy. While ground-based data center networks can be expanded with additional switches and fiber optics, orbital expansion often means new satellites, new links, and even new launches.

The Pengcheng-NSK-1 satellite, launched on September 20, is thus highly representative. Instead of focusing solely on computing power, it integrates 5G NTN, on-board core networks, AI computing, and satellite-ground laser communication into a single 150 kg satellite, with plans to conduct 100 Gbps satellite-ground laser communication experiments. According to Pengcheng Laboratory's plans, this satellite will also validate terminal access, on-board protocol processing, and service transmission.

Chips are now forcing networks to upgrade together.

04

Similar technological paths have already been validated overseas.

The European Space Agency's Φsat-2, launched in 2024, is only 6U but can run multiple AI applications on orbit for tasks such as cloud detection, ship identification, and disaster mapping. The HPE Spaceborne Computer-2 on the International Space Station has also validated commercial servers performing high-performance computing and AI tasks in orbit.

These projects validate a further capability: data can be processed directly on orbit, AI applications can be deployed to satellites, and commercial computing devices used on Earth can also handle some space-based computing tasks.

China's space-based computing projects are no longer solely focused on increasing single-satellite computing power; software management, networking, and high-speed communication are also advancing together. The Santih Computing Constellation is experimenting with multi-satellite collaboration, the Tiansuan Constellation is bringing software environments and resource management into orbit, AIRSAT-02 is improving satellite-ground data transmission capabilities, and the Pengcheng-NSK-1 satellite is compressing communication, computing, and high-speed links into a single node.

These projects come from different institutions with varying approaches and maturity levels, and they do not yet form a complete orbital data center. However, focusing solely on single-satellite computing power can no longer explain why these projects are simultaneously advancing networking, scheduling, software updates, and high-speed communication.

While having a satellite compute faster is important, as the number of computing nodes in orbit grows, system efficiency will increasingly depend on whether these nodes can collaborate stably, whether tasks can be flexibly allocated, and whether data can flow timely.

A single satellite completing an on-orbit computation is fundamentally different in difficulty from dozens or hundreds of satellites collaborating stably over the long term. The latter must also address power consumption, heat dissipation, radiation, link stability, resource scheduling, and operational costs. If any link falls behind, the preceding computing power will be held back.

Thus, what China is sending into space today is not just more powerful computing payloads. It is attempting to organize computing, networking, software, and high-speed data links into a system that can operate sustainably.

Forty years ago, Silicon Valley used the phrase 'the network is the computer' to explain the future on Earth. Today, that phrase is gaining a new version hundreds of kilometers above Earth.