Graph Neural Networks (GNNs) have drawn tremendous attention due to their unique capability to extend Machine Learning (ML) approaches to applications broadly-defined as having unstructured data, especially graphs. Compared with other Machine Learning (ML) modalities, the acceleration of Graph Neural Networks (GNNs) is more challenging due to the irregularity and heterogeneity derived from graph typologies. Existing efforts, however, have focused mainly on handling graphs' irregularity and have not studied their heterogeneity. To this end we propose H-GCN, a PL (Programmable Logic) and AIE (AI Engine) based hybrid accelerator that leverages the emerging heterogeneity of Xilinx Versal Adaptive Compute Acceleration Platforms (ACAPs) to achieve high-performance GNN inference. In particular, H-GCN partitions each graph into three subgraphs based on its inherent heterogeneity, and processes them using PL and AIE, respectively. To further improve performance, we explore the sparsity support of AIE and develop an efficient density-aware method to automatically map tiles of sparse matrix-matrix multiplication (SpMM) onto the systolic tensor array. Compared with state-of-the-art GCN accelerators, H-GCN achieves, on average, speedups of 1.1~2.3X.
翻译:与其它机器学习(ML)模式相比,图形神经网络的加速更具挑战性,因为来自图形类型学的不规则性和异质性。然而,现有的努力主要侧重于处理图形的不规则性,没有研究其异质性。为此,我们提议H-GCN、一个基于可编程逻辑的PL(可编程逻辑)和AIE(AI引擎)的混合加速器,利用Xilinx Versal兼容加速平台(ACAPs)的新兴异质性来达到高性能GNN的推断。特别是,H-GCN分区的每个图以其固有的异质性为基础,分为三个子图,并分别使用PL和AIE进行加工。为了进一步改进性能,我们探索AIE(可编程逻辑)和AIE(AI引擎)的混合加速器,利用 Xlinx Versive兼容性加速平台(AAPs)的新兴异质性能性,并开发一个高效的GMMAS-S-S-S-S-SDRO-S-S-S-SQRODMAR-S-S-S-SAL-Sl-Sl-S-Sl-S-S-Sl-S-SQ-Sl-Sl-SQ-SQ-SQ-Sl-Sl-Sl-S-S-S-S-S-Sl-S-S-Sl-S-S-S-S-S-S-S-SQ-S-S-S-Sl-S-S-S-S-S-Sl-S-S-S-S-S-S-M-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-S-