EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Researching the CNN Collaborative Inference Mechanism for Heterogeneous Edge Devices

Jian Wang, Chong Chen iD, Shiwei Li, Chaoyong Wang iD, Xianzhi Cao, Liusong Yang

Sensors · 27.06.2024 · 24 (13) · 4176

DOI: 10.3390/s24134176 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 27.06.2024. Verfasst von Jian Wang und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer and Control Engineering, Northeast Forestry University, Harbin 150040, China. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Jian Wang, Special Fund for Forestry Scientific Research in the Public Interest.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Convolutional Neural Networks (CNNs) have been widely applied in various edge computing devices based on intelligent sensors. However, due to the high computational demands of CNN tasks, the limited computing resources of edge intelligent terminal devices, and significant architectural differences among these devices, it is challenging for edge devices to independently execute inference tasks locally. Collaborative inference among edge terminal devices can effectively utilize idle computing and storage resources and optimize latency characteristics, thus significantly addressing the challenges posed by the computational intensity of CNNs. This paper targets efficient collaborative execution of CNN inference tasks among heterogeneous and resource-constrained edge terminal devices. We propose a pre-partitioning deployment method for CNNs based on critical operator layers, and optimize the system bottleneck latency during pipeline parallelism using data compression, queuing, and “micro-shifting” techniques. Experimental results demonstrate that our method achieves significant acceleration in CNN inference within heterogeneous environments, improving performance by 71.6% compared to existing popular frameworks.

Schlagwörter

Stichwörter der Autor:innen

CNN inferenceedgeheterogeneitymodel partitioningpipeline parallelism

Autor:innen und Institutionen

  • College of Computer and Control Engineering, Northeast Forestry University, Harbin 150040, China.

ORCID-Profile: Chong Chen, Chaoyong Wang

Förderung

  • Jian Wang
  • Special Fund for Forestry Scientific Research in the Public Interest

Zitieren

APA

Wang, J., Chen, C., Li, S., Wang, C., Cao, X., & Yang, L. (2024). Researching the CNN Collaborative Inference Mechanism for Heterogeneous Edge Devices. Sensors, 24(13), 4176. https://doi.org/10.3390/s24134176

Vancouver

Wang J, Chen C, Li S, Wang C, Cao X, Yang L. Researching the CNN Collaborative Inference Mechanism for Heterogeneous Edge Devices. Sensors. 2024;24(13):4176. doi:10.3390/s24134176

BibTeX
@article{Wang2024,
  author = {Wang, Jian and Chen, Chong and Li, Shiwei and Wang, Chaoyong and Cao, Xianzhi and Yang, Liusong},
  title = {Researching the CNN Collaborative Inference Mechanism for Heterogeneous Edge Devices},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {13},
  pages = {4176},
  doi = {10.3390/s24134176}
}
Kostenlos starten →