network-topology

Amazon discusses the evolution of flat datacenter network topologies, from theoretical expander graphs to practical implementations like VL2 and Jellyfish, and current research into Penrose tiling-based designs at AWS.

This paper performs full Jacobian eigendecomposition across production-scale LLMs, revealing a learned spectral gradient from rotation-dominated early layers to symmetric late layers, along with a low-rank bottleneck that compresses perturbations. The results link perturbation propagation and compression to network functional topology.