太空部署大模型算力是否划算？研究发现推理可行，训练难竞争地面数据中心。
The Cost and Network Limits of Space-Based AI Compute
- 用激光星间链路构建星载网状网络，替代地面Clos架构
- 轨道算力在推理场景下可行，但训练顶级大模型成本过高
- 适合关注太空计算基础设施的科研与航天机构
本文评估了在低地球轨道（LEO）部署大规模AI数据中心是否可成为地面设施的成本有效替代方案。分析涵盖发射成本、电力生成、冷却、辐射暴露及大气再入等维度，以及算力-网络性能表现。关键区别在于从地面Clos网络转向基于激光星间链路的星载网状网络。通过分叉带宽、分叉强度及类屋顶模型评估，结果显示：尽管轨道推理具备可行性，但在轨道上训练前沿规模的大语言模型（LLM）仍难以与地面数据中心竞争。
原文摘要 · Abstract (English)
This paper evaluates whether large-scale AI data centers deployed in low-Earth orbit (LEO) could become a cost-effective alternative to terrestrial facilities. The analysis compares orbital and ground-based systems across launch cost, power generation, cooling, radiation exposure, and atmospheric reentry, as well as compute-network performance. A key distinction is the shift from terrestrial Clos networks to space-based mesh networks using laser inter-satellite links. Using bisection bandwidth, bisection intensity, and roofline-style models, we show that while LEO-based inference may be feasible, training frontier-scale LLMs in orbit is unlikely to be competitive with terrestrial data centers.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。