Metraipolis › archive › dc_0edbf0d3ca67

Low-Latency MoE Inference in Space Data Centers: A Wireless-Power-Transfer Perspective — arXiv:2610.30923

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.30923
License (as archived)
CC0-1.0
Archived
2026-10-25T08:34:43.520Z
SHA-256 of the archived text
5dccd2821df85e4365b7f5bbf88721a33003da10a348aaad8c1160a4fbd32462
Stable URL of this page
https://metraipolis.com/documento/dc_0edbf0d3ca67

How to cite

Low-Latency MoE Inference in Space Data Centers: A Wireless-Power-Transfer Perspective — arXiv:2610.30923 — arXiv (en), https://arxiv.org/abs/2610.30923 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0edbf0d3ca67 (Metraipolis archive dc_0edbf0d3ca67, sha256 5dccd2821df85e4365b7f5bbf88721a33003da10a348aaad8c1160a4fbd32462)
BibTeX
@misc{dc0edbf0d3ca67,
  title = {Low-Latency MoE Inference in Space Data Centers: A Wireless-Power-Transfer Perspective — arXiv:2610.30923},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30923},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0edbf0d3ca67; sha256 5dccd2821df85e4365b7f5bbf88721a33003da10a348aaad8c1160a4fbd32462},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30923. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Space data centers (SpaceDCs) are emerging as platforms for large language models (LLMs), supported by continuous solar-energy harvesting and high-speed interconnections. The mixture-of-experts (MoE) architecture supports distributed LLM inference by deploying experts across satellites. However, skewed expert activation creates uneven computation power demands despite uniform solar-power supply at the dawn-dusk sun-synchronous orbit (SSO). Consequently, active experts remain power-constrained while harvested power at idle experts is underused. These reduce inference efficiency and increase latency. To address this supply-demand mismatch, we propose the SpaceWPT framework, which uses inter-satellite laser-based wireless power transfer (WPT) to redistribute harvested power. The key design challenge is to balance the computation demands of power-donating and power-receiving satellites under WPT losses: transferring more power accelerates computation at the receiver but leaves less power for the donor. After deriving design principles that simplify WPT control, we characterize the resulting latency tradeoff through a tractable surrogate of average inference latency. Minimizing the surrogate yields a novel two-sided water-filling strategy that determines whether each active expert donates, retains, or receives power according to the water level and its WPT link efficiency. Simulations with Mixtral-8x7B and Phi-3.5-MoE show up to 29.25% lower inference latency than without WPT, with larger gains under more compact formations and higher transfer efficiencies.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Distributed, Parallel, and Cluster Computing (cs.DC) ; Artificial Intelligence (cs.AI); Networking and Internet Architecture (cs.NI).
Pagina del lavoro: https://arxiv.org/abs/2610.30923 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-25T08:31:13.202Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).