SysPilot：基于库性能契约的跨库性能自动调优
SysPilot: Cross-Library Performance Autotuning via Library Performance Contracts
AI 导读
SysPilot 是一个基于库性能契约（LPC）的跨库性能自动调优系统，可从文档和代码合成契约，组合库契约、推导资源冲突假设并通过性能分析验证，选出候选配置。在跨库资源竞争基准测试中，SysPilot 的调优结果最多比 OpenTuner 和 SMAC3 找到的最佳配置再提升 38%，且运行次数少 3 倍，同时契约可跨应用复用。SysPilot 已开源。
正文 · 原文
Abstract:Modern scientific and machine-learning applications compose many independently developed libraries. When these libraries run together, their independently managed runtimes can oversubscribe shared resources: using the same CPU cores, fighting over GPUs, or contending for memory bandwidth. Avoiding such contention requires careful configuration across library boundaries. Existing autotuners do not directly investigate what resources are shared between libraries, which runtime signals correlate with interference. As a result, they are inefficient for finding optimal configurations to mitigate contention.
This paper presents SysPilot, a cross-library performance autotuning system based on Library Performance Contract (LPC). A Library Performance Contract captures a library's resource demands, assumptions, controls, and observability signals in LPCL, a specification language for describing library resource requirements. SysPilot synthesizes contracts from documentation and code artifacts, composes contracts from the libraries used by an application, derives resource-conflict hypotheses, profiles the application to validate those hypotheses, and selects candidate configurations. This design makes the Library Performance Contract a reusable unit of performance knowledge: a contract for a library can be reused across many applications that depend on that library. We evaluate SysPilot on a benchmark suite of cross-library contention scenarios. The results show that SysPilot matches the performance of autotuners that rely on application-specific manual information while retaining cross-application reuse. Compared with general-purpose autotuners such as OpenTuner and SMAC3, SysPilot improves benchmark performance metrics by up to an additional 38% over the best configuration found by these autotuners, while using 3x fewer runs.
| Comments: | SysPilot is open source at this https URL |
| Subjects: | Performance (cs.PF); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.06078 [cs.PF] |
| (or arXiv:2611.06078v1 [cs.PF] for this version) | |
| https://doi.org/10.48550/arXiv.2611.06078 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yineng Yan [view email]
[v1]
Wed, 4 Nov 2026 07:20:45 UTC (274 KB)
来源：arXiv cs.AI · arxiv.org