Spectra：为 AI 训练调度并行光电路交换机

View PDF HTML (experimental)

Abstract:The rapid growth of AI training has dramatically increased datacenter traffic demand and energy consumption, which has motivated renewed interest in optical circuit switches (OCSes) as a high-bandwidth, energy-efficient alternative for AI fabrics. Deploying multiple parallel OCSes is a leading alternative. However, efficiently scheduling time-varying traffic matrices across parallel optical switches with non-negligible reconfiguration delays remains an open challenge.
We consider the problem of scheduling a single AI traffic demand matrix $D$ over $s$ parallel OCSes while minimizing the makespan under reconfiguration delay $\delta$. Our algorithm Spectra relies on a three-step approach: Decompose $D$ into a minimal set of weighted permutations; Schedule these permutations across parallel switches using load-aware assignment; then Equalize the imbalanced loads on the switches via controlled permutation splitting. Evaluated on realistic AI training workloads (GPT model and Qwen MoE expert routing) as well as standard benchmarks, Spectra vastly outperforms a baseline based on state-of-the-art algorithms, reducing schedule makespan by an average factor of $1.4\times$ on GPT AI workloads, $2.7\times$ on MoE AI workloads, and $2.4\times$ on standard benchmarks. Further, the makespans achieved by Spectra consistently approach newly derived lower bounds.

Comments:	Accepted to IEEE/Optica Journal of Lightwave Technology
Subjects:	Networking and Internet Architecture (cs.NI); Artificial Intelligence (cs.AI)
Cite as:	arXiv:2603.07373 [cs.NI]
	(or arXiv:2603.07373v2 [cs.NI] for this version)
	https://doi.org/10.48550/arXiv.2603.07373 arXiv-issued DOI via DataCite

Submission history

From: Isaac Keslassy [view email]
[v1] Sat, 7 Mar 2026 22:57:12 UTC (2,219 KB)
[v2] Fri, 9 Oct 2026 20:57:00 UTC (2,824 KB)