EUVIMEDEUVIMED entdecken
Crossref

M^2 -DFU: Multi-Mode Dataflow Architecture for Adaptive and High-Efficiency Data Processing

Wenming Li, Zhihua Fan, Haibin Wu, Kunming Zhang, Yu Liu, Haoran Tong, Xiaochun Ye, Dongrui Fan

ACM Transactions on Computer Systems · 20.08.2026

DOI: 10.1145/3834567 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Computer Systems, online erschienen am 20.08.2026. Verfasst von Wenming Li und 7 weiteren Autor:innen. Die Arbeit nennt 51 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The impending end of Moore’s Law and Dennard Scaling has prompted a shift in both industry and academia from general-purpose processors (e.g., CPUs and GPGPUs) toward specialized solutions. However, the diverse hardware requirements of edge computing make it challenging for existing specialized architectures to adapt to rapidly evolving algorithms, often prioritizing efficiency at the expense of generality. Our approach leverages the inherent advantages of software-defined dataflow mechanisms and integrates reconfigurable SIMD into the architecture to achieve high efficiency and flexibility in hardware for edge computing, surpassing ASIC architectures. In this work, we propose M ^2 -DFU (Multi-Mode Dataflow Unit), a flexible, highly efficient architecture with configurable multiple execution modes. First, we identify general and specific requirements for various edge applications, including convolutional neural networks (CNNs), digital signal processing (DSP), and typical scientific computing. Next, we design a flexible architecture with high configurability across multiple modes. Efficient and reconfigurable microarchitectural optimizations are implemented to achieve multilevel parallelism and ensure high hardware utilization. Compared to NVIDIA GPUs, M ^2 -DFU achieves an average improvement of up to 6.8 \times in energy efficiency for typical edge computing applications.

Zitieren

APA

Li, W., Fan, Z., Wu, H., Zhang, K., Liu, Y., Tong, H., Ye, X., & Fan, D. (2026). M^2 -DFU: Multi-Mode Dataflow Architecture for Adaptive and High-Efficiency Data Processing. ACM Transactions on Computer Systems. https://doi.org/10.1145/3834567

Vancouver

Li W, Fan Z, Wu H, Zhang K, Liu Y, Tong H, et al. M^2 -DFU: Multi-Mode Dataflow Architecture for Adaptive and High-Efficiency Data Processing. ACM Transactions on Computer Systems. 2026. doi:10.1145/3834567

BibTeX
@article{Li2026,
  author = {Li, Wenming and Fan, Zhihua and Wu, Haibin and Zhang, Kunming and Liu, Yu and Tong, Haoran and Ye, Xiaochun and Fan, Dongrui},
  title = {M^2 -DFU: Multi-Mode Dataflow Architecture for Adaptive and High-Efficiency Data Processing},
  journal = {ACM Transactions on Computer Systems},
  year = {2026},
  doi = {10.1145/3834567}
}
Kostenlos starten →