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

TL;DR

GCL-Sampler uses graph contrastive learning to automatically discover kernel similarities in GPU simulation traces, enabling faster simulations with high accuracy by capturing structural and semantic execution features.

Contribution

It introduces a novel graph contrastive learning framework that automatically learns high-dimensional kernel similarities from trace graphs for GPU simulation sampling.

Findings

Achieves 258.94x average speedup with 0.37% error.

Outperforms state-of-the-art methods like PKA, Sieve, STEM+ROOT.

Effectively captures structural and semantic properties of GPU kernels.

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%).