# Architectural Implications of Graph Neural Networks

arXiv: https://arxiv.org/abs/2009.00804 · HTML: https://mlsanity.com/paper/2009.00804/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Sep 2020.

## Built on by (most cited first, 12 of 12)

- [Parallel and Distributed Graph Neural Networks: An In-Depth Concurrency Analysis](https://mlsanity.com/paper/2205.09702/) (2022)
- [A Comprehensive Survey on Distributed Training of Graph Neural Networks](https://mlsanity.com/paper/2211.05368/) (2022)
- [Revisiting Bayesian Autoencoders with MCMC](https://mlsanity.com/paper/2104.05915/) (2021)
- [SMASH: Sparse Matrix Atomic Scratchpad Hashing](https://mlsanity.com/paper/2105.14156/) (2021)
- [Characterizing the Efficiency of Graph Neural Network Frameworks with a Magnifying Glass](https://mlsanity.com/paper/2211.03021/) (2022)
- [STAG: Enabling Low Latency and Low Staleness of GNN-based Services with Dynamic Graphs](https://mlsanity.com/paper/2309.15875/) (2023)
- [Characterizing and Understanding HGNN Training on GPUs](https://mlsanity.com/paper/2407.11790/) (2024)
- [ProGNNosis: A Data-driven Model to Predict GNN Computation Time Using Graph Metrics](https://mlsanity.com/paper/2206.08258/) (2022)
- [Towards Real-Time Temporal Graph Learning](https://mlsanity.com/paper/2210.04114/) (2022)
- [gSuite: A Flexible and Framework Independent Benchmark Suite for Graph Neural Network Inference on GPUs](https://mlsanity.com/paper/2210.11601/) (2022)
- [Accelerating Generic Graph Neural Networks via Architecture, Compiler, Partition Method Co-Design](https://mlsanity.com/paper/2308.08174/) (2023)
- [SiHGNN: Leveraging Properties of Semantic Graphs for Efficient HGNN Acceleration](https://mlsanity.com/paper/2408.15089/) (2024)

## Abstract (arXiv)

Graph neural networks (GNN) represent an emerging line of deep learning models that operate on graph structures. It is becoming more and more popular due to its high accuracy achieved in many graph-related tasks. However, GNN is not as well understood in the system and architecture community as its counterparts such as multi-layer perceptrons and convolutional neural networks. This work tries to introduce the GNN to our community. In contrast to prior work that only presents characterizations of GCNs, our work covers a large portion of the varieties for GNN workloads based on a general GNN description framework. By constructing the models on top of two widely-used libraries, we characterize the GNN computation at inference stage concerning general-purpose and application-specific architectures and hope our work can foster more system and architecture research for GNNs.