AI Compute & Infrastructure / Edge AI Inference & Distributed Architectures / Metro Edge vs. Centralized Cloud Latency Benchmarks