[论文] The Cost of Saying Hello: An Empirical Study of Handshake Overhead in ...

论文概要

研究领域: ML 作者: Pranav Varshney 发布时间: 2026-10-24 arXiv: 2610.30151

中文摘要

每次 AI 智能体交互都从「握手」开始：发起方要先定位对方、弄清它能做什么、建立会话，然后才能发出第一条有效消息。协议综述常把手销开销列为效率指标，却从没有人给出实测数据。本文首次对三大智能体协议（MCP 的 2025 initialize 与 2026 discover 两个阶段、A2A、ACP）做分阶段的实证基准：基于各官方 SDK，逐项测量往返次数、分阶段延迟、线上字节与 token 成本，覆盖冷启动、热重连、能力不匹配、元数据退化四类场景（每格 N=50，含回环与 50ms RTT 模拟）。结论：往返次数主导握手成本——旧版 MCP 握手需 3 次往返、50ms RTT 下耗时 230ms；A2A 仅需 1 次往返、61ms。但往返不是全部账单：当 LLM 需要阅读对方元数据时，成本随对方能力清单规模线性增长（每项能力 24–107 token），一个 200 项能力的对手方会在智能体读到任务之前就吃掉 200k 上下文窗口的 11%。MCP 2026 修订版把握手从 3 次往返砍到 2 次、配合 era pinning 热重连可降为 0——协议的自我演进恰好佐证了本文测得的成本。全部 harness、原始数据与分析已开源。

原文摘要

Every interaction between AI agents begins with a handshake: the initiating agent must locate its counterpart, learn what it can do, and establish a session before the first useful message can be sent. Surveys of agent communication protocols name handshake overhead as an efficiency criterion, yet none report measurements. This paper presents the first empirical, phase-decomposed benchmark of the handshake in three major agent protocols: MCP (in both its 2025 initialize and 2026 discover eras), A2A, and ACP. Using each protocol's official SDK, we instrument every request for round trips, per-phase latency, wire bytes, and token cost across cold-start, warm-repeat, capability-mismatch, and degraded-metadata scenarios (N=50 per cell, on loopback and under an emulated 50 ms RTT). Round-trip count dominates handshake cost. A legacy MCP handshake takes three round trips and 230 ms at 50 ms RTT; A2A needs one round trip and 61.0 ms. Round trips are not the whole bill: when an LLM reads the counterpart's metadata, handshake cost scales with the counterpart's capability inventory at 24-107 tokens per capability, so a 200-capability counterpart spends 11% of a 200k-token context window before the agent reads its own task. MCP's own 2026 revision cuts its handshake from three round trips to two and, with era pinning, to zero on warm reconnects, protocol evolution that corroborates the cost we measure. We release the full harness, raw data, and analysis at this https URL.

---

*自动采集于 2026-10-24*

#论文 #arXiv #ML #小凯