用AI代理自动调优异构存储系统，性能提升超5倍，延迟降低近90%。
StorageXTuner: An LLM Agent-Driven Automatic Tuning Framework for Heterogeneous Storage Systems
- 分四类智能体协同工作，动态探索配置空间。
- 在多个数据库上实现最高575%吞吐提升，尾延迟降88%。
- 适合需要跨系统自动调优的工程师和系统研发者。
自动配置存储系统难度大：参数空间庞大，且受工作负载、部署环境和版本变化影响。传统启发式与机器学习调优方法通常依赖特定系统、需手动集成，且在环境变化时性能下降。近期基于大模型的方法虽有进展，但多将调优视为单次、系统专属任务，限制了跨系统复用、探索能力与验证强度。本文提出 StorageXTuner，一个面向异构存储引擎的 LLM 代理驱动自动调优框架。该框架通过四个智能体分离职责：执行器（沙箱基准测试）、提取器（性能摘要）、搜索器（基于洞察的配置探索）与反思器（洞察生成与管理）。设计结合洞察驱动的树搜索与分层记忆机制，促进经实证验证的优化见解积累，并使用轻量级检查器防止不安全操作。我们实现了原型，在 RocksDB、LevelDB、CacheLib 与 MySQL InnoDB 上，采用 YCSB、MixGraph 与 TPC-H/C 进行评估。相比默认配置和 ELMo-Tune，StorageXTuner 最高实现 575% 和 111% 的吞吐提升，尾延迟（p99）最多降低 88% 和 56%，且收敛所需尝试次数更少。
原文摘要 · Abstract (English)
Automatically configuring storage systems is hard: parameter spaces are large and conditions vary across workloads, deployments, and versions. Heuristic and ML tuners are often system specific, require manual glue, and degrade under changes. Recent LLM-based approaches help but usually treat tuning as a single-shot, system-specific task, which limits cross-system reuse, constrains exploration, and weakens validation. We present StorageXTuner, an LLM agent-driven auto-tuning framework for heterogeneous storage engines. StorageXTuner separates concerns across four agents - Executor (sandboxed benchmarking), Extractor (performance digest), Searcher (insight-guided configuration exploration), and Reflector (insight generation and management). The design couples an insight-driven tree search with layered memory that promotes empirically validated insights and employs lightweight checkers to guard against unsafe actions. We implement a prototype and evaluate it on RocksDB, LevelDB, CacheLib, and MySQL InnoDB with YCSB, MixGraph, and TPC-H/C. Relative to out-of-the-box settings and to ELMo-Tune, StorageXTuner reaches up to 575% and 111% higher throughput, reduces p99 latency by as much as 88% and 56%, and converges with fewer trials.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。