arXiv：cs.AI· Vasilis Kypriotis, Alexander Blasberg, Alexandros Moiras, Dimitrios Skarlatos·· 22 小时前
Agentic Architect：面向 CPU 微架构设计探索与优化的智能体 AI 框架
Agentic Architect: An Agentic AI Framework for Architecture Design Exploration and Optimization
AI 导读
Agentic Architect 是一个结合 LLM 代码演化、周期精确仿真与硬件成本建模的智能体框架，用于 CPU 微架构设计空间探索与优化。在覆盖五个基准套件的 213 条 trace 上，其演化出的预取器在 L1D 和 L2 上分别比最强 SOTA 设计额外提速最高 6.7% 和 2.6%，最佳缓存替换设计比 Mockingjay 提升最高 1.7%。
正文
Abstract:Computer Architecture research has traditionally required the exploration of vast, highly complex design spaces through manual human effort. Large Language Models (LLMs) offer an opportunity to accelerate this process by enabling fast and automated implementation and testing of different architectural mechanisms and their combinations. We introduce Agentic Architect, a framework for agentic CPU microarchitectural design-space exploration and optimization. The Agentic Architect combines LLM-driven code evolution with cycle-accurate simulation and hardware cost modeling to explore performance, area, and power tradeoffs. Starting from a seed design, the LLM is prompted to explore the design space by implementing changes, guided by prior evaluation feedback. Each new design is compiled, simulated on training workloads, evaluated against the optimization objective, and constrained by area and power budgets. Its score and diagnostics update the candidate population and guide subsequent iterations.
We evaluate the Agentic Architect framework across prefetching, cache replacement, and branch prediction. On 213 traces spanning five benchmark suites, its evolved prefetchers lead the evaluated L1D and L2 designs, delivering up to 6.7% and 2.6% additional speedup, respectively, over the strongest state-of-the-art design, while the best evolved cache replacement design improves over Mockingjay by up to 1.7%. Moreover, across 191 branch-prediction traces, the evolved branch predictor achieves 97.62% of TAGE-SC-L's geometric mean IPC at 10.58% of its area. Our analysis highlights the importance of seed selection, hardware constraints, and training workloads in shaping search outcomes. Overall, Agentic Architect establishes the first end-to-end framework for agentic CPU microarchitectural design optimization, opening a new direction for architecture research.
| Subjects: | Artificial Intelligence (cs.AI); Hardware Architecture (cs.AR) |
| Cite as: | arXiv:2604.25083 [cs.AI] |
| (or arXiv:2604.25083v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2604.25083 arXiv-issued DOI via DataCite |
Submission history
From: Dimitrios Skarlatos [view email]
[v1]
Tue, 28 Apr 2026 00:31:55 UTC (389 KB)
[v2]
Mon, 12 Oct 2026 05:07:15 UTC (1,448 KB)
来源：arXiv：cs.AI · arxiv.org