A Probabilistic Machine Learning Approach to Scheduling Parallel Loops with Bayesian Optimization

TL;DR

This paper introduces BO FSS, a Bayesian optimization-based automatic tuning method for parallel loop scheduling, which improves performance and robustness over traditional algorithms by modeling execution time with Gaussian processes.

Contribution

It presents a novel Bayesian optimization approach with a locality-aware Gaussian process model to automatically tune FSS parameters for better parallel loop scheduling.

Findings

BO FSS improves execution time by up to 22% over FSS.

BO FSS demonstrates more consistent performance across workloads.

The locality-aware GP model accelerates Bayesian optimization convergence.

Abstract

This paper proposes Bayesian optimization augmented factoring self-scheduling (BO FSS), a new parallel loop scheduling strategy. BO FSS is an automatic tuning variant of the factoring self-scheduling (FSS) algorithm and is based on Bayesian optimization (BO), a black-box optimization algorithm. Its core idea is to automatically tune the internal parameter of FSS by solving an optimization problem using BO. The tuning procedure only requires online execution time measurement of the target loop. In order to apply BO, we model the execution time using two Gaussian process (GP) probabilistic machine learning models. Notably, we propose a locality-aware GP model, which assumes that the temporal locality effect resembles an exponentially decreasing function. By accurately modeling the temporal locality effect, our locality-aware GP model accelerates the convergence of BO. We implemented BO FSS on the GCC implementation of the OpenMP standard and evaluated its performance against other scheduling algorithms. Also, to quantify our method's performance variation on different workloads, or workload-robustness in our terms, we measure the minimax regret. According to the minimax regret, BO FSS shows more consistent performance than other algorithms. Within the considered workloads, BO FSS improves the execution time of FSS by as much as 22% and 5% on average.