SCV-GNN: Sparse Compressed Vector-based Graph Neural Network Aggregation

TL;DR

SCV-GNN introduces a novel sparse compressed vectors format optimized for GNN aggregation, significantly improving speed and reducing memory traffic, thereby enhancing GNN inference efficiency on specialized hardware.

Contribution

The paper proposes the SCV-GNN format with Z-Morton ordering for better data locality, demonstrating scalability and substantial performance gains over traditional sparse formats.

Findings

Achieves nearly 8x speedup over CSC and CSR methods.

Reduces memory traffic by over 3x and 4x respectively.

Enhances GNN inference latency and memory efficiency.

Abstract

Graph neural networks (GNNs) have emerged as a powerful tool to process graph-based data in fields like communication networks, molecular interactions, chemistry, social networks, and neuroscience. GNNs are characterized by the ultra-sparse nature of their adjacency matrix that necessitates the development of dedicated hardware beyond general-purpose sparse matrix multipliers. While there has been extensive research on designing dedicated hardware accelerators for GNNs, few have extensively explored the impact of the sparse storage format on the efficiency of the GNN accelerators. This paper proposes SCV-GNN with the novel sparse compressed vectors (SCV) format optimized for the aggregation operation. We use Z-Morton ordering to derive a data-locality-based computation ordering and partitioning scheme. The paper also presents how the proposed SCV-GNN is scalable on a vector processing system. Experimental results over various datasets show that the proposed method achieves a geometric mean speedup of $7.96\times$ and $7.04\times$ over CSC and CSR aggregation operations, respectively. The proposed method also reduces the memory traffic by a factor of $3.29\times$ and $4.37\times$ over compressed sparse column (CSC) and compressed sparse row (CSR), respectively. Thus, the proposed novel aggregation format reduces the latency and memory access for GNN inference.