Chain-NN: An energy-efficient 1D chain architecture for accelerating deep convolutional neural networks

Shihao Wang, Dajiang Zhou, Xushen Han, Takeshi Yoshimura

研究成果: Conference contribution

27 被引用数 (Scopus)

抄録

Deep convolutional neural networks (CNN) have shown their good performances in many computer vision tasks. However, the high computational complexity of CNN involves a huge amount of data movements between the computational processor core and memory hierarchy which occupies the major of the power consumption. This paper presents Chain-NN, a novel energy-efficient 1D chain architecture for accelerating deep CNNs. Chain-NN consists of the dedicated dual-channel process engines (PE). In Chain-NN, convolutions are done by the 1D systolic primitives composed of a group of adjacent PEs. These systolic primitives, together with the proposed column-wise scan input pattern, can fully reuse input operand to reduce the memory bandwidth requirement for energy saving. Moreover, the 1D chain architecture allows the systolic primitives to be easily reconfigured according to specific CNN parameters with fewer design complexity. The synthesis and layout of Chain-NN is under TSMC 28nm process. It costs 3751k logic gates and 352KB on-chip memory. The results show a 576-PE Chain-NN can be scaled up to 700MHz. This achieves a peak throughput of 806.4GOPS with 567.5mW and is able to accelerate the five convolutional layers in AlexNet at a frame rate of 326.2fps. 1421.0GOPS/W power efficiency is at least 2.5 to 4.1x times better than the state-of-the-art works.

本文言語English
ホスト出版物のタイトルProceedings of the 2017 Design, Automation and Test in Europe, DATE 2017
出版社Institute of Electrical and Electronics Engineers Inc.
ページ1032-1037
ページ数6
ISBN(電子版)9783981537093
DOI
出版ステータスPublished - 2017 5 11
イベント20th Design, Automation and Test in Europe, DATE 2017 - Swisstech, Lausanne, Switzerland
継続期間: 2017 3 272017 3 31

Other

Other20th Design, Automation and Test in Europe, DATE 2017
国/地域Switzerland
CitySwisstech, Lausanne
Period17/3/2717/3/31

ASJC Scopus subject areas

  • コンピュータ ネットワークおよび通信
  • ハードウェアとアーキテクチャ
  • 安全性、リスク、信頼性、品質管理

フィンガープリント

「Chain-NN: An energy-efficient 1D chain architecture for accelerating deep convolutional neural networks」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル