Abstract

Various emerging technologies have been proposed and applied to deep convolutional neural networks (DCNNs) to enhance their execution efficiency and accuracy while reducing the number of parameters. However, previous dataflow-based DCNN accelerators have primarily focused on accelerating the convolution (CONV) layer while reducing the overhead of data movement through specific data reuse strategies to improve performance. When deploying these emerging DCNNs in conventional accelerators, frequent data access from both the external memory and internal processing units of the accelerator can result in significant data movement overhead, leading to decreased acceleration performance. To address these challenges, this paper proposes a novel three-dimensional hybrid optical-electrical network-on-chip (NoC) accelerator called the optical-electronic channel-stationary system (OECS). OECS leverages a channel stationary (CS) calculation mode and stay-at-local data storage strategy to minimize the movement and processing of all data in the local processing element (PE). This strategy reduces data movement between each PE in the accelerator and between the accelerator and external memory, resulting in improved acceleration performance. Simulation results demonstrate that, compared to state-of-the-art accelerators, OECS achieves a 53% improvement in execution speed and saves approximately 44% of energy consumption associated with data movement.

© 2023 Optica Publishing Group

Universal wavelength reuse mechanism for optical networks-on-chip based on a cooperative game

Hongyu Yang, Yiyuan Xie, Tingting Song, Ye Su, Bocheng Liu, Junxiong Chai, Xiao Jiang, Li Dai, and Jing Pang
J. Opt. Commun. Netw. 15(6) 367-380 (2023)

On the network design and control of an optical network: interconnecting multiple chips on a wafer

Ziyue Zhang, Didier Colle, Wouter Tavernier, and Mario Pickavet
J. Opt. Commun. Netw. 15(2) 119-132 (2023)

Exploration of a neural-network-based joint method of mapping and wavelength assignment in optical network-on-chip

Hui Li, Yuxiang Niu, and Feiyang Liu
J. Opt. Commun. Netw. 15(9) 600-619 (2023)

Efficient Sharing of Optical Resources in Low-Power Optical Networks-on-Chip

Sebastian Werner, Javier Navaridas, and Mikel Luján
J. Opt. Commun. Netw. 9(5) 364-374 (2017)

Deep-learning-based path computation without routing convergence in optical satellite networks

Yinji Jing, Longteng Yi, Yongli Zhao, Hua Wang, Wei Wang, and Jie Zhang
J. Opt. Commun. Netw. 15(5) 294-303 (2023)