用K点批处理与混合精度缓解平面波DFT的GPU瓶颈

训练下一代机器学习原子间势(MLIP)需要真值:跨多种结构和化学组成,进行数千万到数亿次密度泛函理论(DFT)计算,以得到能量、力和应力。平面波 DFT 以 VASP 作为其主流实现,是生成这类数据的金标准。它很准确……
译文由英文原文机器翻译,术语按行业惯例统一,可在文末展开原文核对。
查看英文原文(Tackling GPU Bottlenecks in Plane-Wave DFT with K-Point Batching and Mixed Precision)
Training a next-generation machine learning interatomic potential (MLIP) requires ground truth: tens to hundreds of millions of density functional theory (DFT)...
Training a next-generation machine learning interatomic potential (MLIP) requires ground truth: tens to hundreds of millions of density functional theory (DFT) calculations of energies, forces, and stresses across diverse structures and chemistries. Plane-wave DFT, with VASP as its dominant implementation, is the gold standard for generating that data. It is accurate…
正文保留英文原文,标题与摘要由机器翻译。
本文由采集器自动同步
阅读原文 ↗
原文发布于 NVIDIA Developer(Tackling GPU Bottlenecks in Plane-Wave DFT with K-Point Batching and Mixed Precision),标题、摘要与正文由机器翻译,术语按行业惯例统一,未经人工逐句校订。