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

A study of massively parallel high-performance graph partitioning using continuous relaxation-based algorithms

Research Project

Project/Area Number 19K20280
Research Category

Grant-in-Aid for Early-Career Scientists

Allocation TypeMulti-year Fund
Review Section Basic Section 60090:High performance computing-related
Research InstitutionUniversity of Tsukuba

Principal Investigator

Futamura Yasunori  筑波大学, システム情報系, 助教 (30736210)

Project Period (FY) 2019-04-01 – 2022-03-31
Project Status Completed (Fiscal Year 2021)
Budget Amount *help
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2021: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2020: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
Fiscal Year 2019: ¥780,000 (Direct Cost: ¥600,000、Indirect Cost: ¥180,000)
Keywordsグラフ解析 / 固有値解析 / グラフアルゴリズム / 固有値計算 / 並列計算 / グラフ分割 / 大規模グラフ / 数値線形代数 / 固有値解法
Outline of Research at the Start

グラフ分割問題は科学技術計算における偏微分方程式求解や、ソーシャルネットワーク解析、並列計算の負荷バランス最適化等に広く応用されている。本研究では離散最適化問題であるグラフ分割問題を連続緩和した問題の大域的最適化に基づく連続緩和型アルゴリズムの高度化に取り組み、超大規模グラフの高速・高精度なグラフ分割を可能とする分散並列ソフトウェアを開発する。

Outline of Final Research Achievements

In this study, we have developed a fast parallel algorithm for solving graph partitioning problems for large graphs. This study targets algorithms based on global optimization of the graph partitioning problem with continuous relaxation. As a notable achievement, we developed a fast eigenspace approximation method using a geodesic distance-based projection, and showed that it has the comparable accuracy and parallel performance as mt-Metis, a standard parallel graph partitioning software, and at the same time has reproducible parallelism, which mt-Metis does not have. Our paper was accepted to IEEE HPEC 2021 and received the Outstanding Paper Award.

Academic Significance and Societal Importance of the Research Achievements

グラフ分割問題は、科学技術計算における偏微分方程式求解や、ソーシャルネットワーク解析、バイオインフォマティクスなど、多様な分野で応用されている。特に高性能計算分野では、タスク割当における負荷バランスと通信コストの最適化に用いられている。新たに提案したアルゴリズムは、高性能計算分野で重点的に研究されている計算カーネルが主要部となっており、それらの高性能実装を活用することで、さらなる高速化が期待できる。近年、解析結果の再現性が重要視されているが、提案法は再現可能な並列性をもつための各種応用分野での再現性保証に資する。本研究成果が波及することにより、科学技術計算や人工知能技術の高度化に寄与する。

Report

(4 results)
  • 2021 Annual Research Report   Final Research Report ( PDF )
  • 2020 Research-status Report
  • 2019 Research-status Report
  • Research Products

    (13 results)

All 2022 2021 2020 2019 Other

All Int'l Joint Research (1 results) Journal Article (6 results) (of which Int'l Joint Research: 1 results,  Peer Reviewed: 6 results,  Open Access: 2 results) Presentation (6 results) (of which Int'l Joint Research: 4 results)

  • [Int'l Joint Research] University of Wuppertal(ドイツ)

    • Related Report
      2021 Annual Research Report
  • [Journal Article] Flexible subspace iteration with moments for an effective contour integration‐based eigensolver2022

    • Author(s)
      Huber Sarah、Futamura Yasunori、Galgon Martin、Imakura Akira、Lang Bruno、Sakurai Tetsuya
    • Journal Title

      Numerical Linear Algebra with Applications

      Volume: online first Issue: 6 Pages: 1-14

    • DOI

      10.1002/nla.2447

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] Structural analysis based on unsupervised learning: Search for a characteristic low-dimensional space by local structures in atomistic simulations2022

    • Author(s)
      Tamura Ryo、Matsuda Momo、Lin Jianbo、Futamura Yasunori、Sakurai Tetsuya、Miyazaki Tsuyoshi
    • Journal Title

      Physical Review B

      Volume: 105 Issue: 7 Pages: 1-18

    • DOI

      10.1103/physrevb.105.075107

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Spectral Graph Partitioning Using Geodesic Distance-based Projection2021

    • Author(s)
      Futamura Yasunori、Wakaki Ryota、Sakurai Tetsuya
    • Journal Title

      2021 IEEE High Performance Extreme Computing Conference (HPEC)

      Volume: proceedings Pages: 1-7

    • DOI

      10.1109/hpec49654.2021.9622831

    • Related Report
      2021 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Efficient Contour Integral-based Eigenvalue Computation Using an Iterative Linear Solver with Shift-Invert Preconditioning2021

    • Author(s)
      Futamura Yasunori、Sakurai Tetsuya
    • Journal Title

      HPC Asia 2021: The International Conference on High Performance Computing in Asia-Pacific Region

      Volume: 該当なし Pages: 90-99

    • DOI

      10.1145/3432261.3432269

    • Related Report
      2020 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] Efficient Implementation of a Dimensionality Reduction Method Using a Complex Moment-Based Subspace2021

    • Author(s)
      Takahiro Yano, Yasunori Futamura, Akira Imakura, Tetsuya Sakurai
    • Journal Title

      In: Proceedings of HPC Asia 2021: The International Conference on High Performance Computing in Asia-Pacific Region

      Volume: 1 Pages: 83-89

    • DOI

      10.1145/3432261.3432267

    • Related Report
      2020 Research-status Report
    • Peer Reviewed / Open Access
  • [Journal Article] Spectral anomaly detection in large graphs using a complex moment-based eigenvalue solver2020

    • Author(s)
      Yasunori Futamura, Xiucai Ye, Akira Imakura, Tetsuya Sakurai
    • Journal Title

      ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering

      Volume: 6 Issue: 2 Pages: 04020010-04020010

    • DOI

      10.1061/ajrua6.0001054

    • Related Report
      2019 Research-status Report
    • Peer Reviewed
  • [Presentation] Spectral Graph Partitioning Using Geodesic Distance-based Projection2021

    • Author(s)
      Futamura Yasunori、Wakaki Ryota、Sakurai Tetsuya
    • Organizer
      2021 IEEE High Performance Extreme Computing Conference (HPEC)
    • Related Report
      2021 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Efficient Contour Integral-based Eigenvalue Computation Using an Iterative Linear Solver with Shift-Invert Preconditioning2021

    • Author(s)
      Futamura Yasunori、Sakurai Tetsuya
    • Organizer
      HPC Asia 2021: The International Conference on High Performance Computing in Asia-Pacific Region
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Efficient Implementation of a Dimensionality Reduction Method Using a Complex Moment-Based Subspace2021

    • Author(s)
      Yano Takahiro、Futamura Yasunori、Imakura Akira、Sakurai Tetsuya
    • Organizer
      HPC Asia 2021: The International Conference on High Performance Computing in Asia-Pacific Region
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] Futamura Yasunori、Sakurai Tetsuya2021

    • Author(s)
      Parallel Contour Integral-Based Eigenvalue Computation using a Block Krylov Linear Solver with Shift-Invert Preconditioning
    • Organizer
      SIAM Conference on Computational Science and Engineering (CSE21)
    • Related Report
      2020 Research-status Report
    • Int'l Joint Research
  • [Presentation] 構造解析向け超並列固有値ソルバの大規模実応用モデルにおける性能評価2019

    • Author(s)
      二村保徳, 櫻井鉄也, 畠澤作二郎
    • Organizer
      日本応用数理学会2019年度年会
    • Related Report
      2019 Research-status Report
  • [Presentation] Spectral nested dissection for large-scale sparse matrix factorization2019

    • Author(s)
      Yasunori Futamura, Tetsuya Sakurai
    • Organizer
      2019 Mini-Workshop on Computational Science
    • Related Report
      2019 Research-status Report

URL: 

Published: 2019-04-18   Modified: 2023-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi