Leveraging LLMs to Automate Energy-Aware Refactoring of Parallel Scientific Codes
This addresses energy efficiency in parallel scientific computing, offering a novel automated refactoring framework that is incremental in improving existing LLM-based methods.
The paper tackles the problem of generating energy-efficient parallel scientific codes using LLMs, proposing LASSI-EE, which achieves an average 29% expected energy reduction with a single run and up to 48% with multiple runs, outperforming vanilla LLM prompting by 2.8x.
While large language models (LLMs) are increasingly used for generating parallel scientific codes, most efforts emphasize functional correctness, often overlooking performance, especially energy efficiency. We propose LASSI-EE, an automated LLM-based refactoring framework that generates energy-efficient parallel codes through a multi-stage, iterative approach integrating runtime power profiling, energy-aware prompting, self-correcting feedback loops, and an LLM-as-a-Judge agent for automated screening of code solutions. We introduce energy-reduction@k, a novel metric that quantifies expected energy reduction when generating k code candidates and selecting the most energy-efficient, enabling systematic evaluation of multi-attempt generation strategies. Evaluating 20 HeCBench applications and two miniApps on NVIDIA A100 and AMD MI100 GPUs, a single run (k=1) with LASSI-EE delivers refactored parallel codes with an average 29% expected energy reduction at an 81% pass rate, representing a 2.8x improvement over vanilla LLM prompting. Multiple runs (k=3) achieve an average 48% expected energy reduction at a 97% pass rate. These results are consistent across devices, demonstrating LASSI-EE's effectiveness across diverse hardware architectures.