Showing cs.NIShow all
2 papers · 1 filter
cs.NI2026
RingStitch: Demand-Aware Optical Stitching for Fragmented TPU Clusters
Huiru Ao, Fan Yang, Binglei Wang +2
Large-scale AI training clusters increasingly use optical circuit switching (OCS) to reconfigure rack-level interconnects and create elastic accelerator slices. In multi-tenant TPU…
cs.NI2025
REACH: Reinforcement Learning for Efficient Allocation in Community and Heterogeneous Networks
Zhiwei Yu, Chengze Du, Heng Xu +3
Community GPU platforms are emerging as a cost-effective and democratized alternative to centralized GPU clusters for AI workloads, aggregating idle consumer GPUs from globally dis…