IFP Energies Nouvelles, 1 et 4 avenue de Bois-Préau, F-92852 Rueil-Malmaison Cedex, France.
J Comput Chem. 2012 Dec 15;33(32):2581-9. doi: 10.1002/jcc.23096. Epub 2012 Aug 20.
We present a way to improve the performance of the electronic structure Vienna Ab initio Simulation Package (VASP) program. We show that high-performance computers equipped with graphics processing units (GPUs) as accelerators may reduce drastically the computation time when offloading these sections to the graphic chips. The procedure consists of (i) profiling the performance of the code to isolate the time-consuming parts, (ii) rewriting these so that the algorithms become better-suited for the chosen graphic accelerator, and (iii) optimizing memory traffic between the host computer and the GPU accelerator. We chose to accelerate VASP with NVIDIA GPU using CUDA. We compare the GPU and original versions of VASP by evaluating the Davidson and RMM-DIIS algorithms on chemical systems of up to 1100 atoms. In these tests, the total time is reduced by a factor between 3 and 8 when running on n (CPU core + GPU) compared to n CPU cores only, without any accuracy loss.
我们提出了一种改进电子结构维也纳从头算模拟包（VASP）程序性能的方法。我们表明，配备图形处理单元（GPU）作为加速器的高性能计算机可以通过将这些部分卸载到图形芯片上来大幅缩短计算时间。该过程包括：（i）分析代码的性能以隔离耗时部分，（ii）重写这些部分，使算法更适合所选的图形加速器，以及（iii）优化主机计算机和 GPU 加速器之间的内存流量。我们选择使用 NVIDIA GPU 来加速 VASP，并使用 CUDA。我们通过在化学系统上评估多达 1100 个原子的 Davidson 和 RMM-DIIS 算法来比较 GPU 和原始 VASP 版本。在这些测试中，与仅使用 n 个 CPU 内核相比，在 n（CPU 内核+GPU）上运行时，总时间减少了 3 到 8 倍，而没有任何精度损失。
