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

Multi-Layer In-Memory Computing and Its Application for GNNs

Research Project

Project/Area Number 22K21284
Research Category

Grant-in-Aid for Research Activity Start-up

Allocation TypeMulti-year Fund
Review Section 1001:Information science, computer engineering, and related fields
Research InstitutionTokyo Institute of Technology (2023)
Keio University (2022)

Principal Investigator

Fujiki Daichi  東京工業大学, 科学技術創成研究院, 准教授 (60963254)

Project Period (FY) 2022-08-31 – 2024-03-31
Project Status Completed (Fiscal Year 2023)
Budget Amount *help
¥2,860,000 (Direct Cost: ¥2,200,000、Indirect Cost: ¥660,000)
Fiscal Year 2023: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2022: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Keywords計算機アーキテクチャ / インメモリ計算 / メモリアーキテクチャ / キャッシュ / PIM
Outline of Research at the Start

インメモリ計算というメモリの中で直接計算を行う技術が、データ移動に係る膨大なコストを低減させる効果があるとして注目を集めている。本研究では、多様なメモリを持つ現代のコンピューティングシステムが、このインメモリ計算技術によってどう変貌するか探求する。そのため、複数のインメモリ計算基盤に着目しながら、GNNなどの動的なアプリケーションの効果的な実行方法を模索し、複数レイヤインメモリ計算技術の有用性を検証する。

Outline of Final Research Achievements

This research addressed architectural challenges in multi-layer in-memory computing (MLIMP) systems. To actively leverage the trade-offs between heterogeneous memories in the memory hierarchy, we devised task scheduling and performance prediction methods targeting machine learning workloads such as GNN and data-parallel workloads.

Furthermore, we discovered the potential of MLIMP to solve locality utilization issues in memory-centric computing. We achieved memory access locality exploitation, which was difficult in single-level in-memory computing systems, by utilizing multi-layer in-memory computing that leverages performance tradeoffs between memory hierarchies. We introduced the concept of "view" to guarantee input/output coherence and defined a cache coherence protocol extension to enable view reuse.

Academic Significance and Societal Importance of the Research Achievements

本研究の学術的意義は、複数階層インメモリコンピューティングという新しい計算機アーキテクチャの提案と、その上で機械学習やデータ並列処理を効率化するスケジューリング・性能予測手法を提案した点にある。また、コヒーレンスに関する課題を新たに発見し、有用なプロトコル拡張を定義したことは高く評価されている。

社会的意義としては、本研究の成果がGNNなどのデータインテンシブな計算の高速化に繋がり、創薬や金融分野におけるシミュレーションの効率化に貢献できる点が挙げられる。また、インメモリコンピューティングの汎用化は、将来的に様々な計算処理の高速化に繋がり、省エネルギーな社会の実現にも貢献する可能性がある。

Report

(3 results)
  • 2023 Annual Research Report   Final Research Report ( PDF )
  • 2022 Research-status Report
  • Research Products

    (6 results)

All 2024 2023 2022 Other

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

  • [Int'l Joint Research] ミシガン大学(米国)

    • Related Report
      2023 Annual Research Report
  • [Int'l Joint Research] University of Michigan, Ann Arbor(米国)

    • Related Report
      2022 Research-status Report
  • [Journal Article] OSA-HCIM: On-The-Fly Saliency-Aware Hybrid SRAM CIM with Dynamic Precision Configuration2024

    • Author(s)
      Chen Yung-Chin、Ando Shimpei、Fujiki Daichi、Takamaeda-Yamazaki Shinya、Yoshioka Kentaro
    • Journal Title

      Asia and South Pacific Design Automation Conference (ASP-DAC)

      Volume: 29 Pages: 539-544

    • DOI

      10.1109/asp-dac58780.2024.10473966

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Vector-Processing for Mobile Devices: Benchmark and Analysis2023

    • Author(s)
      Khadem Alireza、Fujiki Daichi、Talati Nishil、Mahlke Scott、Das Reetuparna
    • Journal Title

      IEEE International Symposium on Workload Characterization (IISWC)

      Volume: 24 Pages: 15-27

    • DOI

      10.1109/iiswc59245.2023.00036

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed / Int'l Joint Research
  • [Journal Article] MVC: Enabling Fully Coherent Multi-Data-Views through the Memory Hierarchy with Processing in Memory2023

    • Author(s)
      Fujiki Daichi
    • Journal Title

      IEEE/ACM International Symposium on Microarchitecture (MICRO)

      Volume: 56 Pages: 800-814

    • DOI

      10.1145/3613424.3623784

    • Related Report
      2023 Annual Research Report
    • Peer Reviewed
  • [Journal Article] Multi-Layer In-Memory Processing2022

    • Author(s)
      Fujiki Daichi、Khadem Alireza、Mahlke Scott、Das Reetuparna
    • Journal Title

      2022 55th IEEE/ACM International Symposium on Microarchitecture (MICRO)

      Volume: 55 Pages: 920-936

    • DOI

      10.1109/micro56248.2022.00068

    • Related Report
      2022 Research-status Report
    • Peer Reviewed / Open Access / Int'l Joint Research

URL: 

Published: 2022-09-01   Modified: 2025-01-30  

Information User Guide FAQ News Terms of Use Attribution of KAKENHI

Powered by NII kakenhi