[2603.00551] GCL-Sampler: Discovering Kernel Similarity for Sampled GPU Simulation via Graph Contrastive Learning

Computer Science > Performance arXiv:2603.00551 (cs) [Submitted on 28 Feb 2026] Title:GCL-Sampler: Discovering Kernel Similarity for Sampled GPU Simulation via Graph Contrastive Learning Authors:Jiaqi Wang, Jingwei Sun, Jiyu Luo, Han Li, Guangzhong Sun View a PDF of the paper titled GCL-Sampler: Discovering Kernel Similarity for Sampled GPU Simulation via Graph Contrastive Learning, by Jiaqi Wang and 4 other authors View PDF HTML (experimental) Abstract: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%). Subjects: Performance (cs...