A near CXL memory processing architecture for distributed graph neural network inference and training

doi:10.1080/09540091.2026.2650981

A near CXL memory processing architecture for distributed graph neural network inference and training

2026-05-23

0

PRE

AI

DOI：10.1080/09540091.2026.2650981

Original

Original request for help

Save

Abstract

En 中文

Distributed Graph Neural Networks (GNNs) require efficient handling of both fine-grained memory accesses and cross memory-device communication, particularly when scaling to large graphs. However, existing acceleration solutions fail to adequately address low bandwidth utilization and scalability across different models and graph sizes. In this paper, we present OptGNN, a scalable heterogeneous distributed architecture tailored for GNNs. OptGNN addresses the challenges of fine-grained memory access by incorporating a near-memory processing mechanism, which improves internal bandwidth utilization. To optimize external communication, we introduce data packing and scheduling strategies that enhance cross memory-device data transfer efficiency. OptGNN achieves 5.7x performance improvement over baseline distributed GNN acceleration methods and 1.29x performance improvement over SOTA distributed GNN acceleration architecture CLAY. Additionally, the system is designed to support various GNN models and large-scale graphs while ensuring load balancing and high hardware utilization.

Keywords:

Near-data processing

graph accelerator

graph neural networks

distributed GNN

A near CXL memory processing architecture for distributed graph neural network inference and training

Abstract

Journal

Organization