DynamiGraph: A Specialized, Runtime-Aware FPGA Overlay for Ultra Low-Latency GNN Inference on Edge Devices
Micromachines · 10.07.2026 · 17 (7) · 824
DOI: 10.3390/mi17070824 ↗Kurz gefasst
Fachartikel in der Zeitschrift Micromachines, online erschienen am 10.07.2026. Verfasst von Haoran Sun und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Airspace Science and Engineering, Shandong University, Weihai 264209, China. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Graph Neural Networks (GNNs) have become essential for analyzing graph-structured data, yet their deployment on resource-constrained edge devices is severely limited by high computational complexity and irregular memory access patterns. Here, we introduce DynamiGraph, a specialized FPGA-based overlay accelerator engineered for ultra-low-latency GNN inference in edge computing scenarios. Unlike general-purpose accelerators that incur high resource overhead to support a broad range of operators, DynamiGraph adopts a streamlined architecture focusing exclusively on essential General Matrix Multiplication (GEMM) and Sparse–Dense Matrix Multiplication (SpDMM) kernels. We implement a hardware-native runtime optimization mechanism that dynamically exploits graph sparsity via an edge-centric execution flow, eliminating redundant computations without requiring complex static preprocessing. Experimental results on an AXU2CGA edge platform demonstrate that DynamiGraph achieves sub-millisecond inference latencies on small-scale benchmarks (e.g., Cora) and a peak throughput of 1467 inferences per second. Furthermore, our runtime sparsity exploitation yields over 2000× reductions in floating-point operations compared to dense equivalents. These findings indicate that trading off model generality for architectural specialization and runtime awareness offers an efficient architectural alternative for enabling real-time graph intelligence in power- and bandwidth-limited edge environments.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Airspace Science and Engineering, Shandong University, Weihai 264209, China.
ORCID-Profile: Haoran Sun
Zitieren
APA
Sun, H., & Liang, L. (2026). DynamiGraph: A Specialized, Runtime-Aware FPGA Overlay for Ultra Low-Latency GNN Inference on Edge Devices. Micromachines, 17(7), 824. https://doi.org/10.3390/mi17070824
Vancouver
Sun H, Liang L. DynamiGraph: A Specialized, Runtime-Aware FPGA Overlay for Ultra Low-Latency GNN Inference on Edge Devices. Micromachines. 2026;17(7):824. doi:10.3390/mi17070824
BibTeX
@article{Sun2026,
author = {Sun, Haoran and Liang, Likai},
title = {DynamiGraph: A Specialized, Runtime-Aware FPGA Overlay for Ultra Low-Latency GNN Inference on Edge Devices},
journal = {Micromachines},
year = {2026},
volume = {17},
number = {7},
pages = {824},
doi = {10.3390/mi17070824}
}
EUVIMED