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

Reduced-precision formats for high-performance and energy-efficient computations

Research Project

Project/Area Number 16K16062
Research Category

Grant-in-Aid for Young Scientists (B)

Allocation TypeMulti-year Fund
Research Field High performance computing
Research InstitutionTokyo Woman's Christian University (2017-2018)
Institute of Physical and Chemical Research (2016)

Principal Investigator

Mukunoki Daichi  東京女子大学, 理学研究科, 研究員 (90742289)

Project Period (FY) 2016-04-01 – 2019-03-31
Project Status Completed (Fiscal Year 2018)
Budget Amount *help
¥2,210,000 (Direct Cost: ¥1,700,000、Indirect Cost: ¥510,000)
Fiscal Year 2017: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2016: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
KeywordsReduced-precision / Mixed-precision / GPU / 浮動小数点演算 / 混合精度演算 / 浮動小数点 / 計算機システム / 高性能計算 / 浮動小数点表現 / 省電力計算
Outline of Final Research Achievements

This study explored the possibility of reduced-precision formats which have shorter bit length against the IEEE 32/64 bit floating-point for enchance the performance of numerical computations in terms of both computation speed and energy efficiency. We proposed a light-weight implementation of reduced-precision formats on software and demonstrated the performance improvement, in terms of both speed and energy efficiency, on some data-intensive operations on basic linear algebra.

Academic Significance and Societal Importance of the Research Achievements

浮動小数点表現を工夫することによる計算の高性能化は,さまざまな計算処理に応用できる汎用的なアプローチである.本研究の開始時と比べると,現在では本研究と類似のアプローチを試みた研究がいくつか見られるようになり,本研究の着想や議論には意義があったと考えられる.また,本研究期間内には達成できなかった混合精度計算手法への適用や,近年発展が著しいFPGAへの適用が今後期待される.

Report

(4 results)
  • 2018 Annual Research Report   Final Research Report ( PDF )
  • 2017 Research-status Report
  • 2016 Research-status Report
  • Research Products

    (7 results)

All 2019 2018 2017 2016

All Journal Article (1 results) Presentation (6 results) (of which Int'l Joint Research: 2 results)

  • [Journal Article] Level-3 BLASに基づく高精度行列積計算法による高精度かつ再現性のあるBLASルーチンの実装とその最適化2018

    • Author(s)
      椋木大地, 荻田武史, 尾崎克久
    • Journal Title

      情報処理学会研究報告: ハイパフォーマンスコンピューティング

      Volume: 2018-HPC-166 Pages: 1-8

    • Related Report
      2018 Annual Research Report
  • [Presentation] OzBLAS: Accurate and Reproducible BLAS Based on Ozaki Scheme2019

    • Author(s)
      Daichi Mukunoki, Takeshi Ogita, Katsuhisa Ozaki
    • Organizer
      GPU Technology Conference (GTC 2019)
    • Related Report
      2018 Annual Research Report
  • [Presentation] 尾崎スキームに基づく高精度かつ再現性のあるBLASルーチンの実装と自動チューニングの適用2019

    • Author(s)
      椋木大地
    • Organizer
      第22回AT研究会オープンアカデミックセッション(ATOS22)
    • Related Report
      2018 Annual Research Report
  • [Presentation] Reduced-/Extended-precision BLASの実装方法の検討2017

    • Author(s)
      椋木大地, 今村俊幸
    • Organizer
      第5回 大規模並列数値計算技術に関する研究集会
    • Place of Presentation
      理化学研究所計算科学研究機構(兵庫県神戸市)
    • Year and Date
      2017-03-27
    • Related Report
      2016 Research-status Report
  • [Presentation] Implementation Techniques for High Performance BLAS Kernels on Modern GPUs2017

    • Author(s)
      Daichi Mukunoki, Toshiyuki Imamura and Daisuke Takahashi
    • Organizer
      SIAM Conference on Computational Science and Engineering (CSE17)
    • Place of Presentation
      Hilton Atlanta, Atlanta (USA, Georgia)
    • Year and Date
      2017-02-28
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research
  • [Presentation] PascalアーキテクチャGPUにおける線形計算カーネルの実装技術の検討2016

    • Author(s)
      椋木大地, 今村俊幸, 高橋大介
    • Organizer
      GTC Japan 2016
    • Place of Presentation
      ヒルトン東京お台場(東京都港区)
    • Year and Date
      2016-10-05
    • Related Report
      2016 Research-status Report
  • [Presentation] Reduced-Precision Floating-Point Formats on GPUs for High Performance and Energy Efficient Computation2016

    • Author(s)
      Daichi Mukunoki and Toshiyuki Imamura
    • Organizer
      IEEE International Conference on Cluster Computing (Cluster 2016)
    • Place of Presentation
      Palais de China Hotel (Taiwan, Taipei)
    • Year and Date
      2016-09-13
    • Related Report
      2016 Research-status Report
    • Int'l Joint Research

URL: 

Published: 2016-04-21   Modified: 2020-03-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi