Prompting for Performance: Exploring LLMs for Configuring Software

TL;DR

This paper explores the potential of large language models to assist in performance-oriented software configuration by evaluating their ability to identify options, rank configurations, and recommend settings across various systems.

Contribution

It presents a novel investigation into using LLMs for software configuration tasks, highlighting their capabilities and limitations in this context.

Findings

LLMs can align with expert knowledge in some configuration tasks.

Hallucinations and superficial reasoning can occur in LLM outputs.

Preliminary results suggest potential for LLM-assisted software configuration.

Abstract

Software systems usually provide numerous configuration options that can affect performance metrics such as execution time, memory usage, binary size, or bitrate. On the one hand, making informed decisions is challenging and requires domain expertise in options and their combinations. On the other hand, machine learning techniques can search vast configuration spaces, but with a high computational cost, since concrete executions of numerous configurations are required. In this exploratory study, we investigate whether large language models (LLMs) can assist in performance-oriented software configuration through prompts. We evaluate several LLMs on tasks including identifying relevant options, ranking configurations, and recommending performant configurations across various configurable systems, such as compilers, video encoders, and SAT solvers. Our preliminary results reveal both positive abilities and notable limitations: depending on the task and systems, LLMs can well align with expert knowledge, whereas hallucinations or superficial reasoning can emerge in other cases. These findings represent a first step toward systematic evaluations and the design of LLM-based solutions to assist with software configuration.