Poster Type: Research Posters
Author: Connor Vitz (University of Delaware), Aaron Jarmusch (University of Delaware), Sunita Chandrasekaran (University of Delaware)
Abstract: Ozaki-Scheme is an established numerical method for emulating high-precision GEMM on low-precision GPU hardware. Using it means navigating a costly accuracy/performance tradeoff across algorithm variants and their tuning parameters. This work automates that process with a multi-stage pipeline combining rule-based variant selection with an LLM-driven evolutionary tuner (OpenEvolve). New features include a monotonicity-constrained XGBoost classifier predicting whether an accuracy target is reachable before spending GPU time and a throughput-priority mode trading bounded accuracy for up to 106.6% higher GFLOPS. A per-island-search fix stabilizes OpenEvolve convergence, widening the best-valid range from about 45-50K to 61-108K GFLOPS. In the Quantum ESPRESSO AUSURF112 workload, native cuBLAS remains 1.21x faster than the tuned complex-GEMM configuration, and the pipeline reaches 20.6x native cuBLAS on GB10 versus 1.42x on GH200. At the poster session, we will talk through the pipeline, presenting five result panels: throughput mode, feasibility classifier, OE-search convergence, AUSURF benchmark, and GH200/GB10 hardware comparison.
Best Poster Finalist (BP): yes
Poster: PDF
Poster Summary: PDF