arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.22425cs.AIcs.AR
收藏

ArchAgent: 由人工智能驱动的计算机架构发现

ArchAgent: Agentic AI-driven Computer Architecture Discovery

University of California, Berkeley(加州大学伯克利分校);Google(谷歌);Google DeepMind(谷歌DeepMind)
Raghav Gupta, Akanksha Jain, Abraham Gonzalez, Alexander Novikov, Po-Sen Huang, Matej Balog, Marvin Eisenberger, Sergey Shirobokov, Ngân Vũ, Martin Dixon, Boriv… 展开作者

Raghav Gupta, Akanksha Jain, Abraham Gonzalez, Alexander Novikov, Po-Sen Huang, Matej Balog, Marvin Eisenberger, Sergey Shirobokov, Ngân Vũ, Martin Dixon, Borivoje Nikolić, Parthasarathy Ranganathan, Sagar Karandikar

AI总结:

ArchAgent利用人工智能驱动的自动化方法,快速发现并优化计算机架构中的缓存替换策略,显著提升性能并展示出在模拟器中的新发现。

摘要:

敏捷硬件设计流程是满足日益增长的计算需求的关键力量倍增器。最近,具有代理特性的生成式人工智能系统在算法设计、提高代码效率以及在科学领域发现方面展示了显著的进步。通过将这些领域连接起来,我们提出了ArchAgent,这是一个基于AlphaEvolve的自动化计算机架构发现系统。我们展示了ArchAgent能够自动设计和实现最先进的(SoTA)缓存替换策略(构建新的机制/逻辑,而不仅仅是改变参数),在已建立的缓存替换策略设计竞赛范围内广泛进行。在两天内没有人工干预的情况下,ArchAgent生成了一种策略,其在公共多核Google工作负载轨迹上的IPC速度提升比之前的SoTA提高了5.3%。在高度探索的单核SPEC06工作负载上,它在18天内生成了一种策略,其在现有SoTA上的IPC速度提升为0.9%(与现有SoTA报告的类似“胜利边缘”)。ArchAgent在速度上比先前的人工开发SoTA策略快3-5倍。代理流程还能够实现“后硅超专用化”,其中代理调整硬件策略中暴露的运行时可配置参数,以进一步使策略与特定工作负载(混合)对齐。利用这一点,我们展示了在SPEC06工作负载上,相对于先前的SoTA,IPC速度提升了2.4%。最后,我们概述了在代理人工智能时代计算机架构研究的更广泛影响。例如,我们展示了“模拟器逃逸”现象,即代理人工智能流程发现了并利用了流行微架构模拟器中的漏洞——这是由于这些研究工具最初是为一个(现已过去)的世界设计的,当时它们仅由人类以善意方式操作。

英文摘要:

Agile hardware design flows are a critically needed force multiplier to meet the exploding demand for compute. Recently, agentic generative AI systems have demonstrated significant advances in algorithm design, improving code efficiency, and enabling discovery across scientific domains. Bridging these worlds, we present ArchAgent, an automated computer architecture discovery system built on AlphaEvolve. We show ArchAgent's ability to automatically design/implement state-of-the-art (SoTA) cache replacement policies (architecting new mechanisms/logic, not only changing parameters), broadly within the confines of an established cache replacement policy design competition. In two days without human intervention, ArchAgent generated a policy achieving a 5.3% IPC speedup improvement over the prior SoTA on public multi-core Google Workload Traces. On the heavily-explored single-core SPEC06 workloads, it generated a policy in just 18 days showing a 0.9% IPC speedup improvement over the existing SoTA (a similar "winning margin" as reported by the existing SoTA). ArchAgent achieved these gains 3-5x faster than prior human-developed SoTA policies. Agentic flows also enable "post-silicon hyperspecialization" where agents tune runtime-configurable parameters exposed in hardware policies to further align the policies with a specific workload (mix). Exploiting this, we demonstrate a 2.4% IPC speedup improvement over prior SoTA on SPEC06 workloads. Finally, we outline broader implications for computer architecture research in the era of agentic AI. For example, we demonstrate the phenomenon of "simulator escapes", where the agentic AI flow discovered and exploited a loophole in a popular microarchitectural simulator - a consequence of the fact that these research tools were designed for a (now past) world where they were exclusively operated by humans acting in good-faith.

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。

↑