Modern LLM serving is hard to tune because each deployment is a stack of interacting choices: model backend, tensor-parallel shape, prefill/decode split, worker...Modern LLM serving is hard to tune because each deployment is a stack of interacting choices: model backend, tensor-parallel shape, prefill/decode split, worker counts, scheduler settings, routing policy, KV cache behavior, autoscaling thresholds, and topology. Those choices interact across layers, and a local improvement can shift the bottleneck somewhere else. For larger models…

Source