Physics Division, Lawrence Livermore National Laboratory, Livermore, California 94550, USA.
College of Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, USA.
J Chem Phys. 2023 May 28;158(20). doi: 10.1063/5.0147249.
We present a Graphics Processing Unit (GPU)-accelerated version of the real-space SPARC electronic structure code for performing Kohn-Sham density functional theory calculations within the local density and generalized gradient approximations. In particular, we develop a modular math-kernel based implementation for NVIDIA architectures wherein the computationally expensive operations are carried out on the GPUs, with the remainder of the workload retained on the central processing units (CPUs). Using representative bulk and slab examples, we show that relative to CPU-only execution, GPUs enable speedups of up to 6× and 60× in node and core hours, respectively, bringing time to solution down to less than 30 s for a metallic system with over 14 000 electrons and enabling significant reductions in computational resources required for a given wall time.
我们提出了一个图形处理单元 (GPU) 加速的实空间 SPARC 电子结构代码版本，用于在局域密度近似和广义梯度近似下进行 Kohn-Sham 密度泛函理论计算。特别是，我们为 NVIDIA 架构开发了一种基于模块化数学内核的实现方式，其中计算密集型操作在 GPU 上执行，其余工作负载保留在中央处理单元 (CPU) 上。使用代表性的体和片例子，我们表明相对于仅 CPU 执行，GPU 分别实现了高达 6×和 60×的节点和核小时的加速，将解决方案的时间缩短到不到 30 s，对于具有超过 14000 个电子的金属系统，并显著减少了给定计算时间所需的计算资源。
