Hot-Rodding the Browser Engine: Automatic Configuration of JavaScript Compilers

TL;DR

This paper presents an automated machine learning approach to optimize JavaScript compiler parameters, achieving over 35% performance improvements on various benchmarks compared to default settings.

Contribution

It introduces a novel automated configuration method for JavaScript compilers using advanced machine learning and optimization techniques.

Findings

Performance improvements of over 35% on benchmarks

Effective automatic tuning of compiler parameters

Demonstrates significant gains over default configurations

Abstract

Modern software systems in many application areas offer to the user a multitude of parameters, switches and other customisation hooks. Humans tend to have difficulties determining the best configurations for particular applications. Modern optimising compilers are an example of such software systems; their many parameters need to be tuned for optimal performance, but are often left at the default values for convenience. In this work, we automatically determine compiler parameter settings that result in optimised performance for particular applications. Specifically, we apply a state-of-the-art automated parameter configuration procedure based on cutting-edge machine learning and optimisation techniques to two prominent JavaScript compilers and demonstrate that significant performance improvements, more than 35% in some cases, can be achieved over the default parameter settings on a diverse set of benchmarks.