Implementation of GPU-FFT into Planewave Based First Principles Calculation Method

DOI Web Site 参考文献9件 オープンアクセス
  • TOMONO Hidekazu
    Department of Mechanical Engineering Informatics, School of Science and Technology, Meiji University
  • AOKI Masaru
    School of Management, Shizuoka Sangyo University Department of Mechanical Engineering Informatics, School of Science and Technology, Meiji University
  • IITAKA Toshiaki
    RIKEN (The Institute of Physical and Chemical Research)
  • TSUMURAYA Kazuo
    Department of Mechanical Engineering Informatics, School of Science and Technology, Meiji University

説明

We present an implementation of a GPU based FFT routine (Graphics Processing Unit based Fast Fourier Transformation) into a CPU based ab initio periodic DFT (Density Functional Theory) calculation code. The FFT calculation in the CPU based DFT codes is the most time-consuming part; for the 128 silicon system, the fraction of time of a CPU FFT calculation amounts to 0.64 of the whole periodic DFT calculation. The replacement of a double precision FFT in the periodic PWscf code with a single precision FFT gives no appreciable differences in both the numerical total energies and the interatomic forces, guaranteeing the use of a single precision GPU based FFT, CUFFT, for the code. The use of the CUFFT reduces the fraction to 0.20 of the whole PWscf code; the replacement speedups a factor of 2.2 for single CPU system. The use of the multi-CPU system with the GPU FFT accelerates by 2.2f, where f is the acceleration factor of the multi-CPU system. The single precision GPU calculation is implementable in any self-consistent electronic structure code, except for the eigensolver part in the DFT codes.

収録刊行物

参考文献 (9)*注記

もっと見る

関連プロジェクト

もっと見る

キーワード

詳細情報 詳細情報について

問題の指摘

ページトップへ