Record-Remix-Replay: Hierarchical GPU Kernel Optimization using Evolutionary Search

TL;DR

Record-Remix-Replay (R^3) is a hierarchical GPU kernel optimization framework that combines advanced search techniques to automate and accelerate performance tuning across multiple optimization dimensions.

Contribution

The paper introduces R^3, a novel hierarchical optimization framework that integrates LLM-driven evolutionary search, Bayesian optimization, and record-replay techniques for comprehensive GPU kernel tuning.

Findings

R^3 outperforms traditional kernel optimization methods.

It is nearly ten times faster than existing evolutionary search approaches.

R^3 achieves better optimization results for scientific applications.

Abstract

As high-performance computing and AI workloads become increasingly dependent on GPUs, maintaining high performance across rapidly evolving hardware generations has become a major challenge. Developers often spend months tuning scientific applications to fully exploit new architectures, navigating a complex optimization space that spans algorithm design, source implementation, compiler flags and pass sequences, and kernel launch parameters. Existing approaches can effectively search parts of this space in isolation, such as launch configurations or compiler settings, but optimizing across the full space still requires substantial human expertise and iterative manual effort. In this paper, we present Record-Remix-Replay (R^3), a hierarchical optimization framework that combines LLM-driven evolutionary search, Bayesian optimization, and record-replay compilation techniques to efficiently explore GPU kernel optimizations from source-level implementation choices down to compiler pass ordering and runtime configuration. By making candidate evaluation fast and scalable, our approach enables practical end-to-end search over optimization dimensions that are typically treated separately. We show that Record-Remix-Replay can optimize full scientific applications better than traditional approaches over kernel parameters and compiler flags, while also being nearly an order of magnitude faster than modern evolutionary search approaches.