TLV-HGNN: Thinking Like a Vertex for Memory-efficient HGNN Inference

TL;DR

This paper introduces TVL-HGNN, a hardware accelerator for HGNN inference that significantly improves speed and energy efficiency by eliminating redundant memory accesses and intermediate storage through a novel vertex-centric execution paradigm.

Contribution

It proposes a semantics-complete execution paradigm and a vertex grouping technique, enabling memory-efficient and high-performance HGNN inference on reconfigurable hardware.

Findings

Achieves 7.85x speedup over NVIDIA A100

Reduces energy consumption by 98.79%

Outperforms state-of-the-art HGNN accelerators

Abstract

Heterogeneous graph neural networks (HGNNs) excel at processing heterogeneous graph data and are widely applied in critical domains. In HGNN inference, the neighbor aggregation stage is the primary performance determinant, yet it suffers from two major sources of memory inefficiency. First, the commonly adopted per-semantic execution paradigm stores intermediate aggregation results for each semantic prior to semantic fusion, causing substantial memory expansion. Second, the aggregation process incurs extensive redundant memory accesses, including repeated loading of target vertex features across semantics and repeated accesses to shared neighbors due to cross-semantic neighborhood overlap. These inefficiencies severely limit scalability and reduce HGNN inference performance. In this work, we first propose a semantics-complete execution paradigm from a vertex perspective that eliminates per-semantic intermediate storage and redundant target vertex accesses. Building on this paradigm, we design TVL-HGNN, a reconfigurable hardware accelerator optimized for efficient aggregation. In addition, we introduce a vertex grouping technique based on cross-semantic neighborhood overlap, with hardware implementation, to reduce redundant accesses to shared neighbors. Experimental results demonstrate that TVL-HGNN achieves average speedups of 7.85x and 1.41x over the NVIDIA A100 GPU and the state-of-the-art HGNN accelerator HiHGNN, respectively, while reducing energy consumption by 98.79% and 32.61%.