Parallel Implementation of CNN on Multi-FPGA Cluster

Yasuyu Fukushima, Kensuke Iizuka, Hideharu Amano

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

We developed a PYNQ cluster called M-KUBOS that consists of economical Zynq boards that are interconnected through low-cost high-performance GTH serial links. For the software environment, we employed the PYNQ open-source software platform. The PYNQ cluster is anticipated to be a multi-access edge computing (MEC) server for 5G mobile networks. We implemented the ResNet-50 inference accelerator on the PYNQ cluster for image recognition of MEC applications. By estimating the execution time of each ResNet-50 layer, layers of ResNet-50 were divided into four boards so that the execution time of each board would be as equal as possible for efficient pipeline processing. Owing to the PYNQ cluster in which FPGAs were directly connected by high-speed serial links, stream processing without network bottlenecks and pipeline processing between boards were readily realized. The implementation achieved 292 GOPS performance, 75.1 FPS throughput, and 5.15 GOPS/W power efficiency. It achieved 17 times faster speed and 86 times more power efficiency compared to the implementation on the CPU, and 3.8 times more power efficiency compared to the implementation on the GPU.

Original languageEnglish
Title of host publicationProceedings - 2021 IEEE 14th International Symposium on Embedded Multicore/Many-Core Systems-on-Chip, MCSoC 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages77-83
Number of pages7
ISBN (Electronic)9781665438605
DOIs
Publication statusPublished - 2021
Event14th IEEE International Symposium on Embedded Multicore/Many-Core Systems-on-Chip, MCSoC 2021 - Singapore, Singapore
Duration: 2021 Dec 202021 Dec 23

Publication series

NameProceedings - 2021 IEEE 14th International Symposium on Embedded Multicore/Many-Core Systems-on-Chip, MCSoC 2021

Conference

Conference14th IEEE International Symposium on Embedded Multicore/Many-Core Systems-on-Chip, MCSoC 2021
Country/TerritorySingapore
CitySingapore
Period21/12/2021/12/23

Keywords

  • CNN
  • MEC
  • Multi FPGA

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Hardware and Architecture
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Parallel Implementation of CNN on Multi-FPGA Cluster'. Together they form a unique fingerprint.

Cite this