arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2007.13828cs.AR
收藏

GRIP:一种图神经网络加速器结构

GRIP: A Graph Neural Network Accelerator Architecture

Kevin Kiningham, Christopher Re, Philip Levis

摘要:

我们提出了GRIP,一个图神经网络加速器架构设计的低延迟推理。加速GNN具有挑战性,因为它们结合了两种不同类型的计算:算术密集型顶点中心运算和内存密集型边中心运算。GRIP将GNN推理分为一组固定的以边缘和顶点为中心的执行阶段,这些阶段可以在硬件中实现。然后,我们为每个阶段中发现的唯一计算结构专门化每个单元。对于以顶点为中心的阶段,GRIP使用高性能矩阵乘法引擎,并结合专用的内存子系统进行权重计算,以提高重用性。对于以边缘为中心的阶段,GRIP使用多个并行预取和归约引擎来减轻内存访问的不规则性。最后,GRIP支持几个GNN优化,包括一个新的优化称为顶点平铺,增加了重用的重量data.We评估GRIP通过执行合成和布局和路由的28纳米实现能够执行推理的几个广泛使用的GNN模型(GCN,GraphSAGE,G-GCN,和GIN)。在几个基准测试图中,与CPU和GPU基线相比,它将第99百分位延迟分别减少了17倍和23倍,而功耗仅为5 W。

英文摘要:

We present GRIP, a graph neural network accelerator architecture designed for low-latency inference. AcceleratingGNNs is challenging because they combine two distinct types of computation: arithmetic-intensive vertex-centric operations and memory-intensive edge-centric operations. GRIP splits GNN inference into a fixed set of edge- and vertex-centric execution phases that can be implemented in hardware. We then specialize each unit for the unique computational structure found in each phase.For vertex-centric phases, GRIP uses a high performance matrix multiply engine coupled with a dedicated memory subsystem for weights to improve reuse. For edge-centric phases, GRIP use multiple parallel prefetch and reduction engines to alleviate the irregularity in memory accesses. Finally, GRIP supports severalGNN optimizations, including a novel optimization called vertex-tiling which increases the reuse of weight data.We evaluate GRIP by performing synthesis and place and route for a 28nm implementation capable of executing inference for several widely-used GNN models (GCN, GraphSAGE, G-GCN, and GIN). Across several benchmark graphs, it reduces 99th percentile latency by a geometric mean of 17x and 23x compared to a CPU and GPU baseline, respectively, while drawing only 5W.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑