Bayesian Optimization for auto-tuning GPU kernels

TL;DR

This paper applies Bayesian Optimization to efficiently find optimal configurations for GPU kernels, addressing challenges of discrete, constrained, and invalid configurations, and demonstrating superior performance over existing methods.

Contribution

It introduces novel techniques for Bayesian Optimization tailored to GPU kernel tuning, including a contextual variance exploration factor and scalable acquisition functions.

Findings

Our method outperforms existing search strategies significantly.

The approach generalizes well across different test cases.

It effectively handles complex, constrained search spaces.

Abstract

Finding optimal parameter configurations for tunable GPU kernels is a non-trivial exercise for large search spaces, even when automated. This poses an optimization task on a non-convex search space, using an expensive to evaluate function with unknown derivative. These characteristics make a good candidate for Bayesian Optimization, which has not been applied to this problem before. However, the application of Bayesian Optimization to this problem is challenging. We demonstrate how to deal with the rough, discrete, constrained search spaces, containing invalid configurations. We introduce a novel contextual variance exploration factor, as well as new acquisition functions with improved scalability, combined with an informed acquisition function selection mechanism. By comparing the performance of our Bayesian Optimization implementation on various test cases to the existing search strategies in Kernel Tuner, as well as other Bayesian Optimization implementations, we demonstrate that our search strategies generalize well and consistently outperform other search strategies by a wide margin.