# GDR-HGNN: A Heterogeneous Graph Neural Networks Accelerator Frontend with Graph Decoupling and Recoupling

arXiv: https://arxiv.org/abs/2404.04792 · HTML: https://mlsanity.com/paper/2404.04792/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Apr 2024.

## Built on by (most cited first, 4 of 4)

- [RSEA-MVGNN: Multi-View Graph Neural Network with Reliable Structural Enhancement and Aggregation](https://mlsanity.com/paper/2408.07331/) (2024)
- [Characterizing and Understanding HGNN Training on GPUs](https://mlsanity.com/paper/2407.11790/) (2024)
- [DR-CircuitGNN: Training Acceleration of Heterogeneous Circuit Graph Neural Network on GPUs](https://mlsanity.com/paper/2508.16769/) (2025)
- [SiHGNN: Leveraging Properties of Semantic Graphs for Efficient HGNN Acceleration](https://mlsanity.com/paper/2408.15089/) (2024)

## Abstract (arXiv)

Heterogeneous Graph Neural Networks (HGNNs) have broadened the applicability of graph representation learning to heterogeneous graphs. However, the irregular memory access pattern of HGNNs leads to the buffer thrashing issue in HGNN accelerators. In this work, we identify an opportunity to address buffer thrashing in HGNN acceleration through an analysis of the topology of heterogeneous graphs. To harvest this opportunity, we propose a graph restructuring method and map it into a hardware frontend named GDR-HGNN. GDR-HGNN dynamically restructures the graph on the fly to enhance data locality for HGNN accelerators. Experimental results demonstrate that, with the assistance of GDR-HGNN, a leading HGNN accelerator achieves an average speedup of 14.6 times and 1.78 times compared to the state-of-the-art software framework running on A100 GPU and itself, respectively.