OS-R1: Agentic Operating System Kernel Tuning with Reinforcement Learning

2025-08-17

AI Summary

Linux kernel tuning suffers from low efficiency, poor generalizability, and difficulty adapting across diverse workloads. To address these challenges, this paper proposes an agent-based kernel tuning framework that integrates rule-guided reinforcement learning with large language models (LLMs), establishing an interactive configuration exploration environment. We design a two-stage training strategy to enhance the LLM’s reasoning rigor and system performance awareness, and introduce a customized reward function for precise optimization. Experiments demonstrate that our method achieves up to 5.6% higher performance than heuristic baselines across multiple workloads, significantly accelerates convergence, and exhibits high data efficiency and strong cross-scenario generalization. The source code and datasets are publicly available.

Technology Category

Cognitive Modeling & Cognitive Systems: Agent ArchitecturesMultiagent Systems: Multiagent LearningMachine Learning: Kernel Methods

Application Category

Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactionsUser Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendationSearch and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Linux kernel tuning is essential for optimizing operating system (OS) performance. However, existing methods often face challenges in terms of efficiency, scalability, and generalization. This paper introduces OS-R1, an agentic Linux kernel tuning framework powered by rule-based reinforcement learning (RL). By abstracting the kernel configuration space as an RL environment, OS-R1 facilitates efficient exploration by large language models (LLMs) and ensures accurate configuration modifications. Additionally, custom reward functions are designed to enhance reasoning standardization, configuration modification accuracy, and system performance awareness of the LLMs. Furthermore, we propose a two-phase training process that accelerates convergence and minimizes retraining across diverse tuning scenarios. Experimental results show that OS-R1 significantly outperforms existing baseline methods, achieving up to 5.6% performance improvement over heuristic tuning and maintaining high data efficiency. Notably, OS-R1 is adaptable across various real-world applications, demonstrating its potential for practical deployment in diverse environments. Our dataset and code are publicly available at https://github.com/LHY-24/OS-R1.

Problem

• Optimizing Linux kernel performance efficiently and scalably
• Improving generalization of kernel tuning methods
• Enhancing accuracy and efficiency in configuration modifications

Innovation

• Rule-based RL for Linux kernel tuning
• LLM-driven efficient configuration exploration
• Two-phase training for fast convergence