• Search Research Projects
  • Search Researchers
  • How to Use
  1. Back to previous page

Development of efficient arbitrary precision numerical linear computation library optimized for multi-core CPUs

Research Project

Project/Area Number 20K11843
Research Category

Grant-in-Aid for Scientific Research (C)

Allocation TypeMulti-year Fund
Section一般
Review Section Basic Section 60090:High performance computing-related
Research InstitutionShizuoka Institute of Science and Technology

Principal Investigator

Kouya Tomonori  静岡理工科大学, 情報学部, 教授 (80319152)

Project Period (FY) 2020-04-01 – 2023-03-31
Project Status Completed (Fiscal Year 2022)
Budget Amount *help
¥2,860,000 (Direct Cost: ¥2,200,000、Indirect Cost: ¥660,000)
Fiscal Year 2022: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2021: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Fiscal Year 2020: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Keywords多倍長精度浮動小数点演算 / 基本線形計算 / 最適化 / SIMD / OpenMP / 行列乗算 / 尾崎スキーム / 分割統治法 / 多倍長精度演算 / BLAS / AVX2 / 多倍長浮動小数点演算 / 基本線型計算 / 並列化 / マルチコアCPU / 多倍長精度数値計算 / 線型計算
Outline of Research at the Start

本研究は,倍精度浮動小数点数(binary64)を越える多倍長精度をサポートする高精度線形計算ライブラリを,マルチコアCPU向けに最適化して高性能化することを目的とする。特に最適化効果の高い密行列乗算に対し,計算精度に応じた多倍長精度演算を使い分け,無誤差変換技法,並列化技法,分割統治法による演算回数の低減技法を組み合わせ,最大の性能向上を引き出せるソフトウェアライブラリを構築し,既存のライブラリとの性能比較を行って有用性を示す。

Outline of Final Research Achievements

We have developed our original fast basic linear computation library supporting multiple-precision floating-point arithmetic, and already published it as open-source project. That includes multi-component-way fixed-precision arithmetic and optimized routines using AVX2, and multi-digit-way arbitrary precision arithmetic using C-native function call which can reduce overhead due to C++ class. For matrix multiplication, the Ozaki scheme and the Strassen algorithm have been implemented, so we have observed that the Ozaki scheme can be more efficient than Strassen algorithm in middle-precision arithmetic, and already published it as the international proceeding paper.

Academic Significance and Societal Importance of the Research Achievements

我々の開発した多倍長精度最適化基本線形計算ライブラリは,標準的な多倍長精度線形計算ライブラリよりも高速であり,OpenMP, AVX2, 尾崎スキーム,Strassenアルゴリズムといった,あらゆる最適化手法を取り入れている。よって,大規模な悪条件問題をより高速に解くことができ,スーパーコンピュータからコンシューマ向けのノートパソコンまで利用することができるものである。Pythonモジュールとしても実験的に実装しており,深層学習結果の検証や,可視化ツールとの連携も可能であり,敷居の高い多倍長精度計算環境をカジュアルに,高価な商用ソフトウェアを介することなく利用できる。

Report

(4 results)
  • 2022 Annual Research Report   Final Research Report ( PDF )
  • 2021 Research-status Report
  • 2020 Research-status Report
  • Research Products

    (20 results)

All 2023 2022 2021 2020

All Journal Article (8 results) (of which Int'l Joint Research: 2 results,  Peer Reviewed: 7 results,  Open Access: 1 results) Presentation (11 results) Book (1 results)

  • [Journal Article] Acceleration of Multiple Precision Solver for Ill-Conditioned Algebraic Equations with Lower Precision Eigensolver2022

    • Author(s)
      Kouya Tomonori
    • Journal Title

      Lecture Notes in Computer Science (Springer)

      Volume: 13378 Pages: 358-372

    • DOI

      10.1007/978-3-031-10562-3_26

    • ISBN
      9783031105616, 9783031105623
    • Related Report
      2022 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Acceleration of Matrix Multiplication Based on Triple-Double (TD), and Triple-Single (TS) Precision Arithmetic2022

    • Author(s)
      Utsugiri Taiga、Kouya Tomonori
    • Journal Title

      Lecture Notes in Computer Science (Springer)

      Volume: 13378 Pages: 406-423

    • DOI

      10.1007/978-3-031-10562-3_29

    • ISBN
      9783031105616, 9783031105623
    • Related Report
      2022 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Optimization of mixed-precision iterative refinement using parallelized direct methods2022

    • Author(s)
      Kouya Tomonori
    • Journal Title

      2022 International Conference on Engineering and Emerging Technologies (ICEET)

      Volume: 2022 Pages: 1-6

    • DOI

      10.1109/iceet56468.2022.10007230

    • Related Report
      2022 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Acceleration of Multiple Precision Matrix Multiplication Based on Multi-component Floating-Point Arithmetic Using AVX22021

    • Author(s)
      Kouya Tomonori
    • Journal Title

      Lecture Notes in Computer Science

      Volume: 12953 Pages: 202-217

    • DOI

      10.1007/978-3-030-86976-2_14

    • ISBN
      9783030869755, 9783030869762
    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Acceleration of LU decomposition supporting double-double, triple-double, and quadruple-double precision floating-point arithmetic with AVX22021

    • Author(s)
      Kouya Tomonori
    • Journal Title

      2021 IEEE 28th Symposium on Computer Arithmetic (ARITH)

      Volume: 2021 Pages: 54-61

    • DOI

      10.1109/arith51176.2021.00021

    • Related Report
      2021 Research-status Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] AVX2を用いたPythonプログラミング環境における多倍長精度線形計算高速化の試み2021

    • Author(s)
      幸谷智紀
    • Journal Title

      静岡理工科大学紀要

      Volume: 29 Pages: 49-56

    • NAID

      120007164219

    • Related Report
      2021 Research-status Report
  • [Journal Article] Performance Evaluation of Strassen Matrix Multiplication Supporting Triple-Double Precision Floating-Point Arithmetic2020

    • Author(s)
      Kouya Tomonori
    • Journal Title

      ICCSA 2020

      Volume: 12253 Pages: 163-176

    • DOI

      10.1007/978-3-030-58814-4_12

    • ISBN
      9783030588137, 9783030588144
    • Related Report
      2020 Research-status Report
    • Peer Reviewed
  • [Journal Article] Python プログラミング環境における多倍長精度数値計算について2020

    • Author(s)
      幸谷智紀
    • Journal Title

      静岡理工科大学紀要

      Volume: 28 Pages: 23-31

    • NAID

      120006848008

    • Related Report
      2020 Research-status Report
    • Peer Reviewed / Open Access
  • [Presentation] 最適化した多倍長精度基本線形計算ライブラリの開発2023

    • Author(s)
      幸谷智紀,打桐大雅
    • Organizer
      第188回HPC研究会
    • Related Report
      2022 Annual Research Report
  • [Presentation] Chebyshev Proxy Rootfinderを用いた非線形方程式の高精度求解の試み2022

    • Author(s)
      幸谷智紀
    • Organizer
      第185回HPC研究会(SWoPP2022)
    • Related Report
      2022 Annual Research Report
  • [Presentation] マルチコンポーネント型多倍長精度行列乗算の最適化について2022

    • Author(s)
      幸谷智紀,打桐大雅
    • Organizer
      日本応用数理学会2022年度年会 OS
    • Related Report
      2022 Annual Research Report
  • [Presentation] 尾崎スキームを用いた6倍精度(Triple-Double) LU分解の高速化2022

    • Author(s)
      打桐大雅,幸谷智紀
    • Organizer
      第186回HPC研究会
    • Related Report
      2022 Annual Research Report
  • [Presentation] 尾崎スキームを用いたDouble-Double(DD)型行列乗算の高速化とその応用2022

    • Author(s)
      打桐大雅,幸谷智紀
    • Organizer
      第187回HPC研究会
    • Related Report
      2022 Annual Research Report
  • [Presentation] AVX2を用いたマルチコンポーネント型多倍長精度直接法の性能評価2021

    • Author(s)
      幸谷智紀
    • Organizer
      第180回HPC研究会(SWoPP 2021)
    • Related Report
      2021 Research-status Report
  • [Presentation] GPU における 3 倍精度浮動小数点数演算 (Triple-Single) の性能評価2021

    • Author(s)
      打桐大雅, 幸谷智紀
    • Organizer
      電気・電子・情報関係学会 東海支部連合研究発表会
    • Related Report
      2021 Research-status Report
  • [Presentation] コンシューマ向けGPUを用いた3倍精度(Triple-Single)行列積の性能評価2021

    • Author(s)
      打桐大雅, 幸谷智紀
    • Organizer
      第182回HPC研究発表会
    • Related Report
      2021 Research-status Report
  • [Presentation] AVX2を用いたマルチコンポーネント型多倍長精度行列乗算の高速化2021

    • Author(s)
      幸谷智紀
    • Organizer
      情報処理学会第178回HPC研究会
    • Related Report
      2020 Research-status Report
  • [Presentation] 3倍精度演算を用いた高速行列乗算2020

    • Author(s)
      幸谷智紀
    • Organizer
      日本応用数理学会2020年度年会
    • Related Report
      2020 Research-status Report
  • [Presentation] SIMD命令を用いた3倍精度行列乗算の性能評価2020

    • Author(s)
      幸谷智紀
    • Organizer
      情報処理学会第176回HPC研究会 (SWoPP2020)
    • Related Report
      2020 Research-status Report
  • [Book] Python数値計算プログラミング2021

    • Author(s)
      幸谷 智紀
    • Total Pages
      272
    • Publisher
      講談社
    • ISBN
      9784065227350
    • Related Report
      2020 Research-status Report

URL: 

Published: 2020-04-28   Modified: 2024-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi