cs.DCNov 10, 2026

Bursts That Outlast Themselves: Persistent Degradation in Prefix-Cached Agent Serving

Authors: Wen Huang, Jack Liu

Organizations: Microsoft · Independent Researcher

Abstract

LLM agents drive inference engines with closed-loop sessions whose context grows every turn, and serving them economically depends on prefix caching: a hit costs an incremental prefill of about a thousand tokens, a miss recomputes a 10--40K-token context. We show that this makes prefix-cached agent serving metastable. With an identical base workload, a few minutes of extra arrivals leaves a vLLM server in a degraded regime--a ~10% hit ratio, about half the throughput, and a backlog that grows for the remaining 70 minutes--while its no-burst twin stays healthy (9 of 12 pairs; none in the reverse direction). The regime appears at loads the cache-warm server sustains, is absent without prefix caching, and also appears with Qwen3-32B on SWE-bench agent trace shapes and on SGLang, which tolerates larger triggers. Slower service keeps more sessions in the system, their working set outgrows the KV pool, and every turn becomes a recomputation; a first-order workload-balance condition, calibrated independently, brackets when the regime can drain. Getting out is the harder problem: throttling new admissions recovered 1 of 7 matched collapsed runs, while suspending admitted sessions against a capacity-based limit recovered all of them within minutes. A CPU offload tier delays the collapse but does not prevent it. A budgeter that bounds the resident KV demand of active sessions to the pool, with thresholds that need no per-workload tuning, prevented the failure in every run we tested.

Figures & tables

Explore similar work

CardsList
  1. Cache Engineering for Agents

    Oct 20, 2026Tianyu Fu, Ruoxi Wang, Zihan Min +6KV-Cache ManagementMemory Management for LLM Agents

  2. TOPAS: Workflow-Aware Prefix-State Scheduling for Multi-Agent LLM Serving

    Aug 26, 2026Hongqiu Ni, Han Tian, Chi Zhang +2LLM Inference SchedulingLLM Agent Serving