极光系列星载智能机 (Aurora Series Satellite-borne Intelligent Machines)
Zhongke Tiansuan (Comospace)
The Aurora series is a line of space-qualified intelligent computing units designed for on-orbit AI processing and supercomputing, featuring different versions with varying performance and architectures.
Key figurescomparable across the catalogue
- Power
- 0.1 W
Normalised to common units so this product can be compared with others in its category. The supplier's own wording is below.
As published by the supplier
- Product Line
- 极光系列星载智能机 (Aurora Series Satellite-borne Intelligent Machines)
- Operating System
- 极光OS (Aurora OS)
- Software Library
- 极光算子库 (Aurora Operator Library)
- Aurora 1000 Series Status
- On-orbit validated for over 1000 days on multiple satellites
- Aurora 1000-3 Development
- 2023, completed three-machine on-orbit collaborative validation
- Aurora 1000 'Huiyan' Development
- 2024, completed on-orbit inference for space-based large models
- Aurora 2000 Series Architecture
- VPX
- Aurora 2000 Series Status
- Development completed, awaiting on-orbit validation
- Aurora 5000 Series GPU
- Domestic full-size GPU (performance comparable to NVIDIA H100)
- Aurora 5000 Series Cooling
- Fluid cooling
- Aurora 5000 Series AI Accelerator Card Support
- Supports over 10 AI accelerator cards for parallel computing
- Aurora 5000 Series Validation Plan
- Planned for on-orbit validation next year
- Tiansuan Project Target Completion
- Gradually built by 2030
- Tiansuan Project Energy Module (Solar Array)
- 1 square kilometer
- Tiansuan Project Energy Module (Power Output)
- Over 100 MW
- Tiansuan Project Computing Module (AI Computing Power)
- 10 Eops
- Tiansuan Project Computing Module (GPU)
- Domestic GPUs
- Tiansuan Project Communication Module (Laser Communication Devices)
- Hundreds of 100 Gbit devices
- Tiansuan Project Communication Module (Total Communication Capacity)
- Up to 10 Tbps
About
The Aurora series satellite-borne intelligent machines are a core product line from Zhongke Tiansuan, designed to bring supercomputing and AI capabilities to space. The product system includes the Aurora series intelligent machines, supporting module cards, the Aurora OS operating system, and the Aurora operator library. Multiple versions of the Aurora series intelligent machines have been released. The Aurora 1000 series has achieved over 1000 days of on-orbit validation across multiple satellites, demonstrating the feasibility of high-performance AI computing in space. The Aurora 2000 series, utilizing a VPX architecture, has completed development and is awaiting on-orbit validation. The upcoming Aurora 5000 series is planned for on-orbit validation next year and will feature a domestic full-size GPU with performance comparable to NVIDIA H100. This series is designed with multi-backup, high reliability, and high-performance characteristics, incorporating fluid cooling for space environments and supporting parallel computing and stable operation of over 10 AI accelerator cards. The Aurora 1000-3 version was developed in 2023, completing collaborative validation experiments with three machines on orbit, exploring multi-node on-orbit collaborative computing architecture. In 2024, the Aurora 1000 ‘Huiyan’ was launched, successfully performing on-orbit inference for space-based large models. The company aims to build a space supercomputing center with its ‘Tiansuan Project’, which envisions a space-based supercomputing system by 2030, comprising an energy module (1 sq km solar array, 100MW power), a computing module (10 Eops AI computing with domestic GPUs, modular, upgradable), and a communication module (hundreds of 100 Gbit laser communication devices, 10 Tbps total capacity). The Aurora series addresses critical challenges of space supercomputing, including fault tolerance in extreme environments (radiation, thermal cycling, micro-vibrations) and heat dissipation for high-power components in vacuum.
Documentation
No public datasheet yet — request the datasheet / ICD from the supplier.