• 研究課題をさがす
  • 研究者をさがす
  • KAKENの使い方
  1. 課題ページに戻る

2019 年度 実施状況報告書

HDLRuby: a new high productivity hardware description language targeting next generation edge computing architectures for IoT

研究課題

研究課題/領域番号 18K11284
研究機関有明工業高等専門学校

研究代表者

Gauthier Lovic  有明工業高等専門学校, 創造工学科, 准教授 (90535717)

研究分担者 石川 洋平  有明工業高等専門学校, 創造工学科, 准教授 (50435476)
白鳥 則郎  中央大学, 研究開発機構, 機構教授 (60111316)
研究期間 (年度) 2018-04-01 – 2022-03-31
キーワードHDL / HW design / Translation / Evaluation / HW simulation / Edge computing
研究実績の概要

The goal of the research is to design a new hardware description language (HDL), named HDLRuby[1], aiming at improving the design productivity of hardware systems.
The previous year was dedicated to the design of the language core and the tools for generating synthesizable Verilog HDL or VHDL hardware description. The work of this year has been focused on four topics: the FPGA and IC implementation and evaluation of a processor design with HDLRuby as real-life case study; the implementation of several generic library components including final state machines, decoders and memories; the design of a new paradigm for a synthesizable unified implementation-independent communication system that allows to change the communication part of an HW component without requiring any reimplementation; the implementation of an RTL simulator for HDLRuby code.
Preliminary works have also been done for studying the introduction of edge computing support (as part of Internet of Things) in HDLRuby.
The works of the previous year lead to two publications:
[1] Evaluation of the HDLRuby Hardware Description Language by implementing an 8-bit RISC Processor
[2] A parallel implementation of deep neural networks distributed over multiple edge devices

現在までの達成度 (区分)
現在までの達成度 (区分)

2: おおむね順調に進展している

理由

The language was already fully usable in the previous year. This year it has been used for producing FPGA and IC implementations of HW described with HDLRuby, for extending the language with several libraries and a new paradigm, for implementing the simulation engine. The evaluation of the language showed that the produced HW has the same level of quality as standard VHDL RTL code while improving the productivity. This evaluation has been published as [1].
Works have also been done for introducing IoT and AI support in the language but they are still preliminary.

今後の研究の推進方策

We plan to go on evaluating the new communication paradigm and actually implement IoT support in the language. It is also planned to implement a neuronal network library for HDLRuby as an application for evaluating the language on large scale applications.

次年度使用額が生じた理由

It was planned to use all the grant amount for the year.
Unfortunately due the COVID-19 outbreak, the trip and attendance to the ICIAE conference for presenting the paper entitled "Evaluation of the HDLRuby Hardware Description Language by implementing an 8-bit RISC Processor" had to be cancelled.
(link: https://www2.ia-engineers.org/iciae2020/)

  • 研究成果

    (3件)

すべて 2020 2019

すべて 雑誌論文 (2件) (うち国際共著 1件、 査読あり 1件、 オープンアクセス 1件) 学会発表 (1件)

  • [雑誌論文] Evaluation of the HDLRuby Hardware Description Language by implementing an 8-bit RISC Processor2020

    • 著者名/発表者名
      Lovic Gauthier, Yohei Ishikawa
    • 雑誌名

      Proceedings of the 8th IIAE International Conference on Industrial Application Engineering

      巻: 8 ページ: オンライン

    • DOI

      10.12792/iciae2020.010

    • 査読あり / オープンアクセス / 国際共著
  • [雑誌論文] 複数エッジデバイスへのモデル並列による分散型ディープニューラルネットワークの通信の実装2019

    • 著者名/発表者名
      大河 亮, 酒井 凌大, ゴーチェ ロヴィック
    • 雑誌名

      電子学会 制御理論・機械学習技術一般

      巻: 1 ページ: 13-17

  • [学会発表] 複数エッジデバイスへのモデル並列による分散型ディープニューラルネットワークの通信の実装2019

    • 著者名/発表者名
      大河 亮
    • 学会等名
      電子学会 制御理論・機械学習技術一般

URL: 

公開日: 2021-01-27  

サービス概要 検索マニュアル よくある質問 お知らせ 利用規程 科研費による研究の帰属

Powered by NII kakenhi