arXiv · 2610.31941
RADICAL: Reconfiguration-Aware Direct-Connect AI Cluster using Spatial-and-Wavelength Selective Switching
Abstract
We present RADICAL, the silicon-photonic-enabled Reconfiguration-Aware Direct-Connect AI Cluster that integrates a custom spatial-and-wavelength selective switch (SWSS) with the Linux network stack. RADICAL deploys a packaged 4x4x4$λ$ microring-based SWSS into a four-GPU-server cluster using standard SFP28 optical transceivers, and a reconfiguration-aware host-networking interface to demonstrate a direct-connect reconfigurable optical fabric. The SWSS provides dynamic spatial and wavelength routing with 3.2-6.8 dB path loss, average wavelength and spatial crosstalk isolation exceeding 25 dB, and error-free 25 Gbps signal recovery across routing paths. At the system level, RADICAL validates reconfiguration-aware operation through continuous bulk data transfers totaling 6.4 TBytes over 10 minutes. We further demonstrate multitenant AI training across reconfigurable link topologies, including four ResNet-18 instances and six nanoGPT instances, with the latter sustained for more than five hours.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Brett George, Weiyang Wang, Yuyang Wang, Zhenguo Wu, S. C. Mancas, Xiang Meng, Manya Ghobadi, Keren Bergman. 2026-10-22. RADICAL: Reconfiguration-Aware Direct-Connect AI Cluster using Spatial-and-Wavelength Selective Switching. https://arxiv.org/abs/2610.31941
Cite the original work for its findings. Save a collection to share your selection of sources.