Graph Neural Networks (GNNs) have shown significant promise in various domains, such as recommendation systems, bioinformatics, and network analysis. However, the irregularity of graph data poses unique challenges for efficient computation, leading to the development of specialized GNN accelerator architectures that surpass traditional CPU and GPU performance. Despite this, the structural diversity of input graphs results in varying performance across different GNN accelerators, depending on their dataflows. This variability in performance due to differing dataflows and graph properties remains largely unexplored, limiting the adaptability of GNN accelerators. To address this, we propose a data-driven framework for dataflow-aware latency prediction in GNN inference. Our approach involves training regressors to predict the latency of executing specific graphs on particular dataflows, using simulations on synthetic graphs. Experimental results indicate that our regressors can predict the optimal dataflow for a given graph with up to 91.28% accuracy and a Mean Absolute Percentage Error (MAPE) of 3.78%. Additionally, we introduce an online scheduling algorithm that uses these regressors to enhance scheduling decisions. Our experiments demonstrate that this algorithm achieves up to $3.17\times$ speedup in mean completion time and $6.26\times$ speedup in mean execution time compared to the best feasible baseline across all datasets.

翻译：图神经网络（GNNs）在推荐系统、生物信息学和网络分析等多个领域展现出巨大潜力。然而，图数据的不规则性给高效计算带来了独特挑战，这推动了超越传统CPU和GPU性能的专用GNN加速器架构的发展。尽管如此，输入图的结构多样性导致不同GNN加速器（取决于其数据流）的性能表现各异。由于不同数据流和图属性导致的这种性能差异在很大程度上尚未得到充分探索，限制了GNN加速器的适应性。为解决此问题，我们提出了一种数据驱动的框架，用于GNN推理中的数据流感知延迟预测。我们的方法包括训练回归器，通过在合成图上进行仿真，来预测特定图在特定数据流上执行的延迟。实验结果表明，我们的回归器能够以高达91.28%的准确率和3.78%的平均绝对百分比误差（MAPE）为给定图预测最优数据流。此外，我们提出了一种在线调度算法，该算法利用这些回归器来优化调度决策。我们的实验表明，与所有数据集上最佳可行基线相比，该算法在平均完成时间上实现了高达$3.17\times$的加速，在平均执行时间上实现了高达$6.26\times$的加速。