Evaluating compiler optimizations for faster and greener web browsing

Abstract

Context: Web browsers are essential in the digital era, enabling seamless access to information and content. However, their intensive resource usage raises energy efficiency concerns, especially in battery-powered devices. Excessive consumption accelerates battery depletion, shortens device lifespan, and contributes to electronic waste and environmental degradation. Objectives: This study aims to empirically assess the impact of compiler optimization flags on the energy efficiency of modern web browsers, focusing on both Rust- and LLVM-based configurations. Methods: Three widely used browsers, namely Firefox, LibreWolf, and Chromium, were evaluated under two benchmark conditions: one browser-specific and one common across all browsers. For each configuration, runtime, energy consumption, and memory usage were measured to quantify performance gains resulting from different optimization strategies. Results: The experiments show that compiler optimizations substantially improve runtime, energy efficiency, and memory utilization, with maximum gains of 76.39%, 76.37%, and 76.22%, respectively. Firefox achieved the best overall performance in energy and runtime metrics, while Chromium showed superior memory efficiency. Rust-specific optimizations alone produced inconsistent outcomes, but combining them with LLVM optimizations yielded more reliable improvements. Conclusion: Compiler optimizations significantly enhance browser performance and energy efficiency. Achieving consistent gains, however, requires carefully combining optimization levels to balance performance and energy use.