GNNHLS: Evaluating Graph Neural Network Inference via High-Level Synthesis

TL;DR

GNNHLS is an open-source framework that evaluates FPGA-based GNN inference acceleration using High-Level Synthesis, demonstrating significant speedup and energy efficiency improvements over CPU and GPU baselines.

Contribution

It introduces a comprehensive FPGA evaluation framework for GNN inference with optimized HLS kernels and a software stack, enabling systematic performance analysis.

Findings

Achieves up to 50.8x speedup over CPU

Realizes 423x energy reduction compared to CPU

Attains up to 5.16x speedup over GPU

Abstract

With the ever-growing popularity of Graph Neural Networks (GNNs), efficient GNN inference is gaining tremendous attention. Field-Programming Gate Arrays (FPGAs) are a promising execution platform due to their fine-grained parallelism, low-power consumption, reconfigurability, and concurrent execution. Even better, High-Level Synthesis (HLS) tools bridge the gap between the non-trivial FPGA development efforts and rapid emergence of new GNN models. In this paper, we propose GNNHLS, an open-source framework to comprehensively evaluate GNN inference acceleration on FPGAs via HLS, containing a software stack for data generation and baseline deployment, and FPGA implementations of 6 well-tuned GNN HLS kernels. We evaluate GNNHLS on 4 graph datasets with distinct topologies and scales. The results show that GNNHLS achieves up to 50.8x speedup and 423x energy reduction relative to the CPU baselines. Compared with the GPU baselines, GNNHLS achieves up to 5.16x speedup and 74.5x energy reduction.