arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2604.28138cs.OScs.AI
收藏

Crab:一种语义感知的检查点/恢复运行时用于智能体沙盒

Crab: A Semantics-Aware Checkpoint/Restore Runtime for Agent Sandboxes

HKUST(香港科技大学)

Tianyuan Wu, Chaokun Chang, Lunxi Cao, Wei Gao, Wei Wang

AI总结:

Crab通过桥接智能体与操作系统间的语义鸿沟,提升检查点恢复的准确性,减少检查点流量,提高执行效率。

摘要:

自主智能体通过沙盒容器和微虚拟机进行操作,其状态涵盖文件系统、进程和运行时 artifact。检查点和恢复(C/R)该状态对于故障容忍、即时执行、强化学习分支和安全回滚至关重要,但现有方法存在两极分化:应用级恢复保留聊天历史但遗漏操作系统副作用,而每轮完整检查点化则正确性高但成本过高。Crab是一种透明的主机侧运行时,无需修改智能体或 C/R 后端,通过 eBPF 基础的 inspector 分类每个轮次的 OS 可见效果以决定检查点粒度;协调器对齐检查点与轮次边界并重叠 C/R 与 LLM 等待时间;主机范围的引擎在共存的沙盒间调度检查点流量。在 shell 密集和代码修复工作负载上,Crab 将恢复正确性从 8%(仅聊天)提升至 100%,减少检查点流量达 87%,并保持在故障自由执行时间的 1.9% 以内。

英文摘要:

Autonomous agents act through sandboxed containers and microVMs whose state spans filesystems, processes, and runtime artifacts. Checkpoint and restore (C/R) of this state is needed for fault tolerance, spot execution, RL rollout branching, and safe rollback-yet existing approaches fall into two extremes: application-level recovery preserves chat history but misses OS-side effects, while full per-turn checkpointing is correct but too expensive under dense co-location. The root cause is an agent-OS semantic gap: agent frameworks see tool calls but not their OS effects; the OS sees state changes but lacks turn-level context to judge recovery relevance. This gap hides massive sparsity: over 75% of agent turns produce no recovery-relevant state, so most checkpoints are unnecessary. Crab (Checkpoint-and-Restore for Agent SandBoxes) is a transparent host-side runtime that bridges this gap without modifying agents or C/R backends. An eBPF-based inspector classifies each turn's OS-visible effects to decide checkpoint granularity; a coordinator aligns checkpoints with turn boundaries and overlaps C/R with LLM wait time; and a host-scoped engine schedules checkpoint traffic across co-located sandboxes. On shell-intensive and code-repair workloads, Crab raises recovery correctness from 8% (chat-only) to 100%, cuts checkpoint traffic by up to 87%, and stays within 1.9% of fault-free execution time.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑