# ReGraphX: NoC-enabled 3D Heterogeneous ReRAM Architecture for Training Graph Neural Networks

arXiv: https://arxiv.org/abs/2102.07959 · HTML: https://mlsanity.com/paper/2102.07959/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Feb 2021.

## Built on by (most cited first, 5 of 5)

- [GHOST: A Graph Neural Network Accelerator using Silicon Photonics](https://mlsanity.com/paper/2307.01782/) (2023)
- [Efficient Mixed Precision Quantization in Graph Neural Networks](https://mlsanity.com/paper/2505.09361/) (2025)
- [Accelerating Neural Networks for Large Language Models and Graph Processing with Silicon Photonics](https://mlsanity.com/paper/2401.06885/) (2024)
- [COIN: Communication-Aware In-Memory Acceleration for Graph Convolutional Networks](https://mlsanity.com/paper/2205.07311/) (2022)
- [Accelerating Generic Graph Neural Networks via Architecture, Compiler, Partition Method Co-Design](https://mlsanity.com/paper/2308.08174/) (2023)

## Abstract (arXiv)

Graph Neural Network (GNN) is a variant of Deep Neural Networks (DNNs) operating on graphs. However, GNNs are more complex compared to traditional DNNs as they simultaneously exhibit features of both DNN and graph applications. As a result, architectures specifically optimized for either DNNs or graph applications are not suited for GNN training. In this work, we propose a 3D heterogeneous manycore architecture for on-chip GNN training to address this problem. The proposed architecture, ReGraphX, involves heterogeneous ReRAM crossbars to fulfill the disparate requirements of both DNN and graph computations simultaneously. The ReRAM-based architecture is complemented with a multicast-enabled 3D NoC to improve the overall achievable performance. We demonstrate that ReGraphX outperforms conventional GPUs by up to 3.5X (on an average 3X) in terms of execution time, while reducing energy consumption by as much as 11X.