ParEVO: Synthesizing Code for Irregular Data: High-Performance Parallelism through Agentic Evolution

TL;DR

ParEVO is a framework that synthesizes high-performance parallel algorithms for irregular data structures, leveraging curated datasets, fine-tuned models, and evolutionary repair techniques to significantly outperform existing models and match expert human performance.

Contribution

The paper introduces ParEVO, combining a large curated task dataset, specialized fine-tuned models, and an evolutionary code repair agent to improve parallel algorithm synthesis for irregular data.

Findings

Achieves up to 1103x speedup on benchmark tasks.

Outperforms state-of-the-art commercial models on complex irregular graph problems.

Matches expert human performance with up to 4.1x speedup on irregular kernels.

Abstract

The transition from sequential to parallel computing is essential for modern high-performance applications but is hindered by the steep learning curve of concurrent programming. This challenge is magnified for irregular data structures (such as sparse graphs, unbalanced trees, and non-uniform meshes) where static scheduling fails and data dependencies are unpredictable. Current Large Language Models (LLMs) often fail catastrophically on these tasks, generating code plagued by subtle race conditions, deadlocks, and sub-optimal scaling. We bridge this gap with ParEVO, a framework designed to synthesize high-performance parallel algorithms for irregular data. Our contributions include: (1) The Parlay-Instruct Corpus, a curated dataset of 13,820 tasks synthesized via a "Critic-Refine" pipeline that explicitly filters for empirically performant algorithms that effectively utilize Work-Span parallel primitives; (2) specialized DeepSeek, Qwen, and Gemini models fine-tuned to align probabilistic generation with the rigorous semantics of the ParlayLib library; and (3) an Evolutionary Coding Agent (ECA) that improves the "last mile" of correctness by iteratively repairing code using feedback from compilers, dynamic race detectors, and performance profilers. On the ParEval benchmark, ParEVO achieves an average 106x speedup (with a maximum of 1103x) across the suite, and a robust 13.6x speedup specifically on complex irregular graph problems, outperforming state-of-the-art commercial models. Furthermore, our evolutionary approach matches state-of-the-art expert human baselines, achieving up to a 4.1x speedup on specific highly-irregular kernels. Source code and datasets are available at https://github.com/WildAlg/ParEVO.