LLMs as Packagers of HPC Software

TL;DR

This paper explores how large language models can assist in generating and refining complex HPC software package recipes, significantly improving installation success rates through a novel framework called SpackIt.

Contribution

It introduces SpackIt, a comprehensive framework that combines repository analysis, example retrieval, and iterative feedback to enhance LLM-based HPC package recipe generation.

Findings

Installation success increased from 20% to over 80%.

Retrieval and feedback methods significantly improve recipe accuracy.

SpackIt demonstrates practical effectiveness on 308 HPC packages.

Abstract

High performance computing (HPC) software ecosystems are inherently heterogeneous, comprising scientific applications that depend on hundreds of external packages, each with distinct build systems, options, and dependency constraints. Tools such as Spack automate dependency resolution and environment management, but their effectiveness relies on manually written build recipes. As these ecosystems grow, maintaining existing specifications and creating new ones becomes increasingly labor-intensive. While large language models (LLMs) have shown promise in code generation, automatically producing correct and maintainable Spack recipes remains a significant challenge. We present a systematic analysis of how LLMs and context-augmentation methods can assist in the generation of Spack recipes. To this end, we introduce SpackIt, an end-to-end framework that combines repository analysis, retrieval of relevant examples, and iterative refinement through diagnostic feedback. We apply SpackIt to a representative subset of 308 open-source HPC packages to assess its effectiveness and limitations. Our results show that SpackIt increases installation success from 20% in a zero-shot setting to over 80% in its best configuration, demonstrating the value of retrieval and structured feedback for reliable package synthesis.