金牛座：加速对十亿规模图的核外图神经网络推理
Taurus: Accelerating Out-of-Core Graph Neural Network Inference on Billion-Scale Graphs
浏览论文内容
中文总结 AI 辅助
针对十亿规模图的GNN推理难题，提出单机系统金牛座，通过重新制定推理方式，借助多种技术手段，在核外图推理中避免读取浪费，减少相关压力和开销，相比基线性能提升显著。
中文摘要 AI 辅助
图神经网络（GNN）对十亿规模图的推理具有挑战性，因为特征和嵌入的内存占用大，且在核外设置中磁盘I/O成本高。现有分布式GNN系统通信时间和基础设施成本高，基于磁盘的GNN系统主要用于训练，在对整个图推理时存在大量读取浪费。我们提出了金牛座，这是一个用于对不适合内存的图进行GNN推理的单机系统，支持精确全图推理和扇出采样推理。为避免随机和重复的特征收集，金牛座将逐层推理重新制定为基于顺序SSD扫描的以源为中心的广播，并由流水线式GPU-CPU-SSD层次结构、拓扑感知重新排序、待处理消息逐出以及用于高度顶点的GPU驻留存储支持。它还使用非缓冲顺序读取和GPU支持的写入来减少页面缓存污染、主机内存压力和写入开销。在具有多达2.69亿个顶点、40亿条边和514 GiB特征的核外图上，金牛座比最强的逐层基线DGI性能高7至25倍，比顶点基线性能高40至140倍。
英文摘要
Graph Neural Network (GNN) inference on billion-scale graphs is challenging due to the large memory footprint of features and embeddings and high disk I/O costs in out-of-core settings. Existing distributed GNN systems incur high communication times and infrastructure costs, while disk-based GNN systems are primarily tailored to training and experience massive wasted reads during inference on the entire graph. We present Taurus, a single-machine system for GNN inference on graphs that do not fit in RAM, supporting both \textit{exact} full-graph inference and fanout-sampled inference. To avoid random and repeated feature gathers, Taurus reformulates layer-wise inference as source-centric broadcasts over sequential SSD scans, backed by a pipelined GPU-CPU-SSD hierarchy, topology-aware reordering, pending-message eviction, and a GPU-resident store for high-degree vertices. It further uses non-buffered sequential reads and GPU-backed writes to reduce page-cache pollution, host-memory pressure, and write overheads. On out-of-core graphs with up to $269M$ vertices, $4B$ edges, and $514$ GiB of features, Taurus outperforms the strongest layer-wise baseline, DGI, by $7$-$25\times$, and vertex-wise baselines by $40$-$140\times$.
发表机构
- Department of Computational and Data Sciences (CDS), Indian Institute of Science (IISc), Bangalore 560012 India（计算与数据科学系，印度科学院（IISc），班加罗尔）
机构由 AI 辅助整理，请以论文原文为准。
