GCN Accelerator

Graph convolutional network (GCN) accelerators aim to significantly speed up and reduce the energy consumption of GCN inference, a crucial task in processing graph-structured data for various applications. Current research focuses on exploiting the inherent sparsity of deep GCNs through compressed data formats and specialized architectures, often employing techniques like in-memory computing and algorithm-hardware co-design to optimize data locality and minimize communication overheads. These advancements are vital for enabling the deployment of GCNs in resource-constrained environments and scaling their application to larger, more complex datasets, impacting fields like social network analysis, drug discovery, and recommendation systems.

Papers

January 25, 2023

SGCN: Exploiting Compressed-Sparse Features in Deep Graph Convolutional Network Accelerators
Mingi Yoo, Jaeyong Song, Jounghoo Lee, Namhyung Kim, Youngsok Kim, Jinho Lee
Graph Neural Network Deep Learning Architecture Feature Sparsity GCN Accelerator Exploiting Sparsity

June 28, 2022

H-GCN: A Graph Convolutional Network Accelerator on Versal ACAP Architecture
Chengming Zhang, Tong Geng, Anqi Guo, Jiannan Tian, Martin Herbordt, Ang Li, Dingwen Tao
Graph Neural Network Phantom 2D Accelerator GNN Inference GCN Accelerator

May 15, 2022

COIN: Communication-Aware In-Memory Acceleration for Graph Convolutional Networks
Sumit K. Mandal, Gokul Krishnan, A. Alper Goksoy, Gopikrishnan Ravindran Nair, Yu Cao, Umit Y. Ogras
Memory Computing Hardware Architecture GCN Accelerator

March 7, 2022

I-GCN: A Graph Convolutional Network Accelerator with Runtime Locality Enhancement through Islandization
Tong Geng, Chunshu Wu, Yongan Zhang, Cheng Tan, Chenhao Xie, Haoran You, Martin C. Herbordt, Yingyan Lin, Ang Li
Graph Neural Network Hardware Accelerator Locality Sensitive GNN Inference Better on Chip Data GCN Accelerator

March 1, 2022

GROW: A Row-Stationary Sparse-Dense GEMM Accelerator for Memory-Efficient Graph Convolutional Neural Networks
Ranggi Hwang, Minhoo Kang, Jiwon Lee, Dongyun Kam, Youngjoo Lee, Minsoo Rhu
GCN Accelerator Dense Matrix Multiplication

December 22, 2021

GCoD: Graph Convolutional Network Acceleration via Dedicated Algorithm and Accelerator Co-Design
Haoran You, Tong Geng, Yongan Zhang, Ang Li, Yingyan Lin
Graph Neural Network Practical Algorithm GCN Model GCN Based Accelerator Design GCN Accelerator

November 4, 2021

LW-GCN: A Lightweight FPGA-based Graph Convolutional Network Accelerator
Zhuofu Tao, Chen Wu, Yuan Liang, Lei He
Graph Neural Network Lightweight High C2p GCN GNN Inference Dense Matrix Multiplication GCN Accelerator

GCN Accelerator

Papers

SGCN: Exploiting Compressed-Sparse Features in Deep Graph Convolutional Network Accelerators

H-GCN: A Graph Convolutional Network Accelerator on Versal ACAP Architecture

COIN: Communication-Aware In-Memory Acceleration for Graph Convolutional Networks

I-GCN: A Graph Convolutional Network Accelerator with Runtime Locality Enhancement through Islandization

GROW: A Row-Stationary Sparse-Dense GEMM Accelerator for Memory-Efficient Graph Convolutional Neural Networks

GCoD: Graph Convolutional Network Acceleration via Dedicated Algorithm and Accelerator Co-Design

LW-GCN: A Lightweight FPGA-based Graph Convolutional Network Accelerator