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

Supporting performance-aware programming with machine learning techniques

Research Project

Project/Area Number 16H02822
Research Category

Grant-in-Aid for Scientific Research (B)

Allocation TypeSingle-year Grants
Section一般
Research Field High performance computing
Research InstitutionTohoku University

Principal Investigator

Hiroyuki Takizawa  東北大学, サイバーサイエンスセンター, 教授 (70323996)

Co-Investigator(Kenkyū-buntansha) 片桐 孝洋  名古屋大学, 情報基盤センター, 教授 (40345434)
横川 三津夫  神戸大学, 先端融合研究環, 教授 (70358307)
南 一生  国立研究開発法人理化学研究所, 計算科学研究センター, ユニットリーダー (70501998)
Research Collaborator Kobayashi Hiroaki  
Suda Reiji  
Okatani Takayuki  
Egawa Ryusuke  
Ohshima Satoshi  
Project Period (FY) 2016-04-01 – 2019-03-31
Project Status Completed (Fiscal Year 2018)
Budget Amount *help
¥16,640,000 (Direct Cost: ¥12,800,000、Indirect Cost: ¥3,840,000)
Fiscal Year 2018: ¥8,060,000 (Direct Cost: ¥6,200,000、Indirect Cost: ¥1,860,000)
Fiscal Year 2017: ¥5,850,000 (Direct Cost: ¥4,500,000、Indirect Cost: ¥1,350,000)
Fiscal Year 2016: ¥2,730,000 (Direct Cost: ¥2,100,000、Indirect Cost: ¥630,000)
Keywords並列処理 / 性能最適化 / 機械学習 / 自動性能チューニング / 自動チューニング
Outline of Final Research Achievements

This work has demonstrated some case studies of effectively using machine learning techniques for supporting High-Performance Computing (HPC) programming. Various problems in code optimization can be solved by converting the problems to the problems that have already been proven to be solved by machine learning. Moreover, this work clarified the importance of analyzing the target problems in advance of machine learning, because it is unlikely that a sufficient number of training data are available in code optimization problems. Moreover, as well as HPC programming, machine learning also needs knowledge and experiences of human experts. However, in machine learning, the problem is already parameterized, and hence can be solved if sufficiently-high performance is available.

Academic Significance and Societal Importance of the Research Achievements

従前,熟練のプログラマによる知識と経験に基づいて,高性能計算アプリケーションコードが対象計算システム向けに最適化されてきた.しかし,ポストムーア時代の大規模かつ複雑な計算システム向けにコードを最適化する労力は,今後ますます増大することが予想され,そのための人材を確保し続けることは困難である.この問題に対して,本研究では近年注目されている機械学習技術を有効活用することで,熟練のプログラマに求められる性能最適化の労力を大幅に軽減できる可能性を明確に示すことができた.多様な科学技術分野で必要不可欠なツールとなっている高性能計算アプリケーションの開発の効率化は,学術的にも社会的にも意義深い成果である.

Report

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

    (43 results)

All 2019 2018 2017 2016 Other

All Int'l Joint Research (1 results) Journal Article (17 results) (of which Int'l Joint Research: 2 results,  Peer Reviewed: 16 results,  Open Access: 2 results,  Acknowledgement Compliant: 1 results) Presentation (23 results) (of which Int'l Joint Research: 13 results,  Invited: 7 results) Book (2 results)

  • [Int'l Joint Research] ダルムシュタット工科大学(ドイツ)

    • Related Report
      2017 Annual Research Report
  • [Journal Article] The Impacts of Locality and Memory Congestion-aware Thread Mapping on Energy Consumption of Modern NUMA Systems2019

    • Author(s)
      Mulya Agung, Muhammad Alfian Amrizal, Ryusuke Egawa and Hiroyuki Takizawa
    • Journal Title

      2019 IEEE Symposium in Low-Power and High-Speed Chips (COOL CHIPS)

      Volume: -

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Performance Evaluation of Different Implementation Schemes of an Iterative Flow Solver on Modern Vector Machines2019

    • Author(s)
      Kenta Yamaguchi, Takashi Soga, Yoichi Shimomura, Thorsten Reimann, Kazuhiko Komatsu, Ryusuke Egawa, Akihiro Musa, Hiroyuki Takizawa, Hiroaki Kobayashi
    • Journal Title

      Supercomputing Frontiers and Innovations

      Volume: -

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed / Open Access / Int'l Joint Research
  • [Journal Article] A Machine Learning-Based Approach for Selecting SpMV Kernels and Matrix Storage Formats2018

    • Author(s)
      Hang Cui and Shoichi Hirasawa and Hiroaki Kobayashi and Hiroyuki Takizawa
    • Journal Title

      IEICE Transactions on Information and Systems

      Volume: E101.D Issue: 9 Pages: 2307-2314

    • DOI

      10.1587/transinf.2017EDP7176

    • NAID

      130007479730

    • ISSN
      0916-8532, 1745-1361
    • Year and Date
      2018-09-01
    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Investigating the Effects of Dynamic Thread Team Size Adjustment for Irregular Applications2018

    • Author(s)
      Xiong Xiao, Mulya Agung, Muhammad Alfian Amrizal, Ryusuke Egawa and Hiroyuki Takizawa
    • Journal Title

      The Sixth International Symposium on Computing and Networking (CANDAR)

      Volume: - Pages: 76-84

    • DOI

      10.1109/candar.2018.00017

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] A Failure Prediction-Based Adaptive Checkpointing Method with Less Reliance on Temperature Monitoring for HPC Applications2018

    • Author(s)
      Muhammad Alfian Amrizal and Pei Li and Mulya Agung and Ryusuke Egawa and Hiroyuki Takizawa
    • Journal Title

      IEEE International Conference on Cluster Computing (CLUSTER2018)

      Volume: - Pages: 512-523

    • DOI

      10.1109/cluster.2018.00067

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Use of Code Structural Features for Machine Learning to Predict Effective Optimizations2018

    • Author(s)
      Yuki Kawarabatake and Mulya Agung and Kazuhiko Komatsu and Ryusuke Egawa and Hiroyuki Takizawa
    • Journal Title

      2018 IEEE International Parallel & Distributed Processing Symposium Workshops

      Volume: - Pages: 1049-1055

    • DOI

      10.1109/ipdpsw.2018.00163

    • Related Report
      2018 Annual Research Report 2017 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Preconditioner auto-tuning with deep learning for sparse iterative algorithms2018

    • Author(s)
      Kenya Yamada and Takahiro Katagiri and Hiroyuki Takizawa and Kazuo Minami and Mitsuo Yokokawa and Toru Nagai and Masao Ogino
    • Journal Title

      The Sixth International Symposium on Computing and Networking Workshops (CANDARW)

      Volume: - Pages: 257-262

    • DOI

      10.1109/candarw.2018.00055

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Automatic Hyperparameter Tuning of Machine Learning Models under Time Constraints2018

    • Author(s)
      Zhen Wang and Agung Mulya and Ryusuke Egawa and Reiji Suda and Hiroyuki Takizawa
    • Journal Title

      IEEE Big Data 2018 Workshop, The Second International Workshop on Automation in Machine Learning and Big Data (AutoML 2018)

      Volume: - Pages: 4967-4973

    • DOI

      10.1109/bigdata.2018.8622384

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Performance Evaluation of a Vector Supercomputer SX-Aurora TSUBASA2018

    • Author(s)
      K. Komatsu, S. Momose, Y. Isobe, O. Watanabe, A. Musa, M. Yokokawa, T. Aoyama, M. Sato, H. Kobayashi
    • Journal Title

      International Conference for High Performance Computing, Networking, Storage and Analysis (SC ’18)

      Volume: - Pages: 685-696

    • DOI

      10.1109/sc.2018.00057

    • Related Report
      2018 Annual Research Report
    • Peer Reviewed
  • [Journal Article] SX-Aurora TSUBASAにおけるプロセス間通信の性能評価2018

    • Author(s)
      塩月信智, 江川隆輔, 滝沢寛之
    • 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(HPC)

      Volume: 2018-HPC-165(21)

    • Related Report
      2017 Annual Research Report
  • [Journal Article] The Importance of Dynamic Load Balancing among OpenMP Thread Teams for Irregular Workloads2017

    • Author(s)
      Xiong Xiao and Shoichi Hirasawa and Hiroyuki Takizawa and Hiroaki Kobayashi
    • Journal Title

      International Journal of Networking and Computing

      Volume: 7(2) Pages: 529-535

    • DOI

      10.1109/candar.2016.0097

    • Related Report
      2017 Annual Research Report
    • Peer Reviewed
  • [Journal Article] A Customizable Auto-Tuning Scenario with User-Defined Code Transformations2017

    • Author(s)
      Hiroyuki Takizawa and Daichi Sato and Shoichi Hirasawa and Daisuke Takahashi
    • Journal Title

      2017 IEEE International Parallel & Distributed Processing Symposium Workshops

      Volume: - Pages: 1372-1378

    • DOI

      10.1109/ipdpsw.2017.79

    • Related Report
      2017 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Optimizing Energy Consumption on HPC Systems with a Multi-level Checkpointing Mechanism2017

    • Author(s)
      Muhammad Alfian Amrizal and Hiroyuki Takizawa
    • Journal Title

      12th International Conference on Networking, Architecture, and Storage(NAS2017)

      Volume: - Pages: 1-9

    • DOI

      10.1109/nas.2017.8026868

    • Related Report
      2017 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Vectorization-Aware Loop Optimization with User-Defined Code Transformations2017

    • Author(s)
      Hiroyuki Takizawa and Thorsten Reimann and Kazuhiko Komatsu and Takashi Soga and Ryusuke Egawa and Akihiro Musa and Hiroaki Kobayashi
    • Journal Title

      IEEE International Conference on Cluster Computing (CLUSTER2017), Workshop on Re-Emergence of Vector Architectures (REV-A)

      Volume: - Pages: 685-692

    • DOI

      10.1109/cluster.2017.102

    • Related Report
      2017 Annual Research Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] A Memory Congestion-aware MPI Process Placement for Modern NUMA Systems2017

    • Author(s)
      Mulya Agung, Muhammad Alfian Amrizal, Kazuhiko Komatsu, Ryusuke Egawa, and Hiroyuki Takizawa
    • Journal Title

      the 24th IEEE International Conference on High Performance Computing, Data, and Analytics (HiPC2017)

      Volume: - Pages: 152-161

    • DOI

      10.1109/hipc.2017.00026

    • Related Report
      2017 Annual Research Report
    • Peer Reviewed
  • [Journal Article] An Application-Level Incremental Checkpointing Mechanism with Automatic Parameter Tuning2017

    • Author(s)
      Hiroyuki Takizawa and Muhammad Alfian Amrizal and Kazuhiko Komatsu and Ryusuke Egawa
    • Journal Title

      The Fifth International Symposium on Computing and Networking, International Workshop on Legacy HPC Application Migration (LHAM2017)

      Volume: - Pages: 389-394

    • DOI

      10.1109/candar.2017.96

    • Related Report
      2017 Annual Research Report
    • Peer Reviewed
  • [Journal Article] A code selection mechanism using deep learning2016

    • Author(s)
      C. Hang, S. Hirasawa, H. Takizawa, and H. Kobayashi
    • Journal Title

      IEEE 10th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC-16)

      Volume: - Pages: 385-392

    • DOI

      10.1109/mcsoc.2016.46

    • Related Report
      2016 Annual Research Report
    • Peer Reviewed / Open Access / Acknowledgement Compliant
  • [Presentation] An Energy Optimization Method for Hybrid In-Memory Checkpointing2019

    • Author(s)
      Muhammad Alfian Amrizal, Mulya Agung, Ryusuke Egawa, Hiroyuki Takizawa
    • Organizer
      IEEE Symposuim on Low-Power and High-Speed Chips and Systems (COOL Chips 22)
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Toward Auto-tuning of Preconditioners for Sparse Iterative Solvers by Deep Learning2019

    • Author(s)
      Takahiro Katagiri
    • Organizer
      2019 Conference on Advanced Topics and Auto Tuning in High-Performance Scientific Computing (ATAT2019)
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] Auto-tuning of Preconditioners with Deep Learning2019

    • Author(s)
      Takahiro Katagiri, Kenya Yamada
    • Organizer
      SIAM Conference on Computational Science and Engineering (CSE19)
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] スーパーコンピュータにおける性能最適化技術2019

    • Author(s)
      南一生
    • Organizer
      最新アーキテクチャ向けプログラムチューニング技術ワークショップ
    • Related Report
      2018 Annual Research Report
    • Invited
  • [Presentation] Memory First! A performance tuning strategy focusing on memory access patterns2019

    • Author(s)
      Hiroyuki Takizawa, Naoki Ebata, Mulya Agung, Muhammand Alan Amrizal, Ryusuke Egawa, Yoko Isobe, and Ryoji Takaki
    • Organizer
      The 29th Workshop on Sustained Simulation Performance
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] NEC SX-Aurora TSUBASA 向けプログラムチューニング技術2019

    • Author(s)
      滝沢寛之
    • Organizer
      最新アーキテクチャ向けプログラムチューニング技術ワークショップ
    • Related Report
      2018 Annual Research Report
    • Invited
  • [Presentation] Expressing the Differences in Code Optimizations between Intel Knights Landing and NEC SX-ACE Processors2018

    • Author(s)
      Hiroyuki Takizawa, Thorsten Reimann, Kazuhiko Komatsu, Takashi Soga, Ryusuke Egawa, Akihiro Musa, and Hiroaki Kobayashi
    • Organizer
      The 13th World Congress on Computational Mechanics/2nd Pan American Congress on Computational Mechanics
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research
  • [Presentation] SX-Aurora TSUBASAの基本性能および機能の初期評価2018

    • Author(s)
      滝沢寛之
    • Organizer
      SX-Aurora TSUBASAフォーラム
    • Related Report
      2018 Annual Research Report
    • Invited
  • [Presentation] Code Optimization with ppOpen-AT and Preconditioner Selection by Deep Learning2018

    • Author(s)
      Takahiro Katagiri
    • Organizer
      International Symposium on Research and Education of Computational Science (RECS2018)
    • Related Report
      2018 Annual Research Report
    • Int'l Joint Research / Invited
  • [Presentation] ディープラーニングによる数値計算ライブラリのチューニングパラメタの最適化の試み2018

    • Author(s)
      山田賢也、 片桐孝洋、永井亨、荻野正雄
    • Organizer
      第23回計算工学講演会
    • Related Report
      2018 Annual Research Report
  • [Presentation] ディープラーニングを用いた数値計算ライブラリにおける反復解法の前処理選択の検討2018

    • Author(s)
      山田 賢也, 片桐 孝洋, 永井 亨, 荻野 正雄
    • Organizer
      日本応用数理学会 2018年度年会
    • Related Report
      2018 Annual Research Report
  • [Presentation] 建物の地震動応答シミュレーションに現れる前処理付き共役勾配法の並列化2018

    • Author(s)
      後藤啓,横川三津夫,坂敏秀
    • Organizer
      第167回ハイパフォーマンスコンピューティング研究会
    • Related Report
      2018 Annual Research Report
  • [Presentation] Automatic Parameter Tuning of Application-Level Incremental Checkpointing2018

    • Author(s)
      Hiroyuki Takizawa, Muhammad Alfian Amrizal, Kazuhiko Komatsu, and Ryusuke Egawa
    • Organizer
      2018 Conference on Advanced Topics and Auto Tuning in High-Performance Scientific Computing (ATAT in HPSC 2018)
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Towards prediction of effective optimizations in performance engineering2018

    • Author(s)
      Hiroyuki Takizawa
    • Organizer
      The 27th Workshop on Sustained Simulation Performance
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] User-Defined Code Transformation for Separation of Performance-Awareness from Application Codes2018

    • Author(s)
      Hiroyuki Takizawa
    • Organizer
      SIAM Conference on Parallel Processing for Scientific Computing
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] High Precision Computing of Matrix-Matrix Multiplications and a New Approach of Auto-Tuning to Numerical Libraries by Deep Learning2018

    • Author(s)
      Takahiro Katagiri, Shuntaro Ichimura and Kenya Yamada
    • Organizer
      SIAM Conference on Parallel Processing for Scientific Computing (PP) 18
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Expectation to Supercomputer Benchmark from the Viewpoint of Performance Optimization Technology2018

    • Author(s)
      Kazuo Minami
    • Organizer
      SIAM Conference on Parallel Processing for Scientific Computing
    • Related Report
      2017 Annual Research Report
    • Int'l Joint Research
  • [Presentation] 疎行列形状のカラー画像を入力としたディープラーニングによる数値計算ライブラリの自動チューニング方式2017

    • Author(s)
      山田賢也、片桐孝洋、永井亨、荻野 正雄
    • Organizer
      情報処理学会研究報告162-HPC-2017
    • Related Report
      2017 Annual Research Report
  • [Presentation] ディープラーニングによるカラー画像を用いた疎行列反復解法ライブラリの自動チューニング2017

    • Author(s)
      山田賢也、片桐孝洋
    • Organizer
      NVIDIA GPU Technical Conference in Japan 2017 (GTC Japan 2017)、テクニカルセッション
    • Related Report
      2017 Annual Research Report
    • Invited
  • [Presentation] ディープラーニングを用いた数値計算ライブラリの最適実装選択の検討2017

    • Author(s)
      山田賢也、片桐孝洋、永井亨、荻野正雄
    • Organizer
      情報処理学会第79回全国大会
    • Place of Presentation
      名古屋大学東山キャンパス(愛知県名古屋市)
    • Related Report
      2016 Annual Research Report
  • [Presentation] 機械学習を用いたコード変換に関する研究2016

    • Author(s)
      川原畑勇希, 平澤将一, 滝沢寛之, 小林広明
    • Organizer
      平成28年度電気関係学会東北支部連合大会
    • Place of Presentation
      東北工業大学八木山キャンパス(宮城県仙台市)
    • Year and Date
      2016-08-30
    • Related Report
      2016 Annual Research Report
  • [Presentation] STView: An Eclipse Plug-in Tool for Visualizing Program Structures in Fortran Source Codes2016

    • Author(s)
      T. Ohichi,M. Terai, M. Yokokawa, and K. Minami
    • Organizer
      The International Conference for High Performance Computing, Networking, Storage and Analysis (SC2016)
    • Place of Presentation
      Salt Lake City, USA
    • Related Report
      2016 Annual Research Report
    • Int'l Joint Research
  • [Presentation] Making a Legacy Code Auto-Tunable without Messing It Up2016

    • Author(s)
      H. Takizawa, D. Sato, S. Hirasawa, and H. Kobayashi
    • Organizer
      The International Conference for High Performance Computing, Networking, Storage and Analysis (SC2016)
    • Place of Presentation
      Salt Lake City, USA
    • Related Report
      2016 Annual Research Report
    • Int'l Joint Research
  • [Book] The Art of High Performance Computing for Computational Science Vol.22019

    • Author(s)
      Editor; Masaaki Geshi
    • Publisher
      Springer
    • Related Report
      2018 Annual Research Report
  • [Book] 計算科学のためのHPC技術22017

    • Author(s)
      下司雅章 編/南 一生他著
    • Total Pages
      340
    • Publisher
      大阪大学出版会
    • Related Report
      2016 Annual Research Report

URL: 

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

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi