用图对比学习自动发现GPU核函数相似性，实现高速高精度模拟。
GCL-Sampler: Discovering Kernel Similarity for Sampled GPU Simulation via Graph Contrastive Learning
- 基于图对比学习的GCN模型自动挖掘指令与数据依赖关系
- 平均258.94倍加速，误差仅0.37%（优于现有方法）
- 适合需要快速高保真GPU架构模拟的研究者
GPU架构仿真比原生执行慢几个数量级，亟需工作负载采样以获得实用加速。现有方法依赖手工特征，表达能力有限，导致采样过于激进（误差高）或保守（加速受限）。为此，我们提出GCL-Sampler，一种利用关系图卷积网络结合对比学习的采样框架，可从追踪图中自动发现高维核函数相似性。通过将指令序列与数据依赖编码为图嵌入，该方法捕捉程序执行的丰富结构与语义特性，实现高保真度与显著加速。在广泛基准测试中，GCL-Sampler相比全量工作负载实现258.94倍平均加速，误差仅0.37%，优于当前最优方法PKA（129.23x，20.90%）、Sieve（94.90x，4.10%）和STEM+ROOT（56.57x，0.38%）。
原文摘要 · Abstract (English)
GPU architectural simulation is orders of magnitude slower than native execution, necessitating workload sampling for practical speedups. Existing methods rely on hand-crafted features with limited expressiveness, yielding either aggressive sampling with high errors or conservative sampling with constrained speedups. To address these issues, we propose GCL-Sampler, a sampling framework that leverages Relational Graph Convolutional Networks with contrastive learning to automatically discover high-dimensional kernel similarities from trace graphs. By encoding instruction sequences and data dependencies into graph embeddings, GCL-Sampler captures rich structural and semantic properties of program execution, enabling both high fidelity and substantial speedup. Evaluations on extensive benchmarks show that GCL-Sampler achieves 258.94x average speedup against full workload with 0.37% error, outperforming state-of-the-art methods, PKA (129.23x, 20.90%), Sieve (94.90x, 4.10%) and STEM+ROOT (56.57x, 0.38%).
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。