ADE-HGNN: Accelerating HGNNs through Attention Disparity Exploitation

TL;DR

This paper introduces ADE-HGNN, a specialized hardware accelerator for HGNNs that employs attention disparity-based pruning and operation fusion to significantly accelerate inference while maintaining accuracy.

Contribution

It proposes a novel runtime pruning method and execution framework for HGNNs, along with a dedicated hardware accelerator design to improve performance and energy efficiency.

Findings

Achieves 28.21x speedup over NVIDIA T4

Achieves 7.98x speedup over NVIDIA A100

Maintains inference accuracy loss within 1.47%

Abstract

Heterogeneous Graph Neural Networks (HGNNs) have recently demonstrated great power in handling heterogeneous graph data, rendering them widely applied in many critical real-world domains. Most HGNN models leverage attention mechanisms to significantly improvemodel accuracy, albeit at the cost of increased computational complexity and memory bandwidth requirements. Fortunately, the attention disparity from source vertices towards a common target vertex unveils an opportunity to boost the model execution performance by pruning unimportant source vertices during neighbor aggregation. In this study, we commence with a quantitative analysis of the attention disparity in HGNN models, where the importance of different source vertices varies for the same target vertex. To fully exploit this finding for inference acceleration, we propose a runtime pruning method based on min-heap and map it to a dedicated hardware pruner to discard unimportant vertices. Given that the pruning overhead itself is non-negligible and cannot be amortized by conventional staged execution paradigm, an operation-fusion execution fow of HGNNs is introduced to overlap the pruning overhead while harnessing inter-stage parallelism. Finally, we present the design of a novel HGNN accelerator, ADE-HGNN, tailored to support the proposed execution framework. Our experimental results demonstrate that ADE-HGNN achieves an average performance improvement of 28.21x over the NVIDIA GPU T4 platform and 7.98x over the advanced GPU A100, with the inference accuracy loss kept within a negligible range of 0.11%~1.47%. Furthermore, ADE-HGNN significantly reduces energy consumption to 1.97% and 5.37% of the two platforms, respectively.