arXiv AI By Srikanta Datta Tumkur, Jay Iyer, Mehar Simhadri, Sai Pavan Kumar, Sai Kapil Kumar, Ramesh Nampelly

Calibrate, Then Route: A Measured Study of Learned Request Routing for Disaggregated LLM Serving

Read the original on arXiv AI →

The paper evaluates a learned request‑routing policy for disaggregated large‑language‑model serving, where compute‑heavy prefill and memory‑heavy decode stages run on separate GPU pools. Using a discrete‑event simulator and real NVIDIA A40 GPUs, the calibrated router—leveraging prompt length, predicted output length, KV‑cache pressure, and SLO class—outperforms round‑robin, least‑loaded, and length‑based heuristics, achieving the highest mean goodput (0.864) and lowest variance across three mixed, bursty arrival traces. Hardware calibration proves critical, providing a 4.5‑point goodput boost and roughly 40 % of the tail‑latency advantage, and the learned router can match round‑robin performance with one fewer GPU in certain scenarios.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.