As the interest to Graph Neural Networks (GNNs) is growing, the importance of benchmarking and performance characterization studies of GNNs is increasing. So far, we have seen many studies that investigate and present the performance and computational efficiency of GNNs. However, the work done so far has been carried out using a few high-level GNN frameworks. Although these frameworks provide ease of use, they contain too many dependencies to other existing libraries. The layers of implementation details and the dependencies complicate the performance analysis of GNN models that are built on top of these frameworks, especially while using architectural simulators. Furthermore, different approaches on GNN computation are generally overlooked in prior characterization studies, and merely one of the common computational models is evaluated. Based on these shortcomings and needs that we observed, we developed a benchmark suite that is framework independent, supporting versatile computational models, easily configurable and can be used with architectural simulators without additional effort. Our benchmark suite, which we call gSuite, makes use of only hardware vendor's libraries and therefore it is independent of any other frameworks. gSuite enables performing detailed performance characterization studies on GNN Inference using both contemporary GPU profilers and architectural GPU simulators. To illustrate the benefits of our new benchmark suite, we perform a detailed characterization study with a set of well-known GNN models with various datasets; running gSuite both on a real GPU card and a timing-detailed GPU simulator. We also implicate the effect of computational models on performance. We use several evaluation metrics to rigorously measure the performance of GNN computation.

Graph Neural Networks (GNNs) have been taking role in many areas, thanks to their expressive power on graph-structured data. On the other hand, Mobile Ad-Hoc Networks (MANETs) are gaining attention as network technologies have been taken to the 5G level. However, there is no study that evaluates the efficiency of GNNs on MANETs. In this study, we aim to fill this absence by implementing a MANET dataset in a popular GNN framework, i.e., PyTorch Geometric; and show how GNNs can be utilized to analyze the traffic of MANETs. We operate an edge prediction task on the dataset with GraphSAGE (SAG) model, where SAG model tries to predict whether there is a link between two nodes. We construe several evaluation metrics to measure the performance and efficiency of GNNs on MANETs. SAG model showed 82.1 accuracy on average in the experiments.

最近，作为基于图形机器学习的骨干的图形神经网络（GNN）展示了各个域（例如，电子商务）的巨大成功。然而，由于基于高稀疏和不规则的图形操作，GNN的性能通常不令人满意。为此，我们提出，TC-GNN，基于GNN加速框架的第一个GPU张量核心单元（TCU）。核心思想是将“稀疏”GNN计算与“密集”TCU进行调和。具体地，我们对主流GNN计算框架中的稀疏操作进行了深入的分析。我们介绍了一种新颖的稀疏图翻译技术，便于TCU处理稀疏GNN工作量。我们还实现了一个有效的CUDA核心和TCU协作设计，以充分利用GPU资源。我们将TC-GNN与Pytorch框架完全集成，以便于编程。严格的实验在各种GNN型号和数据集设置的最先进的深图库框架上平均显示了1.70倍的加速。

虽然有很多关于图像深度学习的硬件加速研究，但在加速涉及图形的深度学习应用时，有一个相当有利的专注。图的独特特性，例如不规则的内存访问和动态并行性，当算法映射到CPU或GPU时，施加有几个挑战。为了在利用所有可用的稀疏性的同时解决这些挑战，我们提出了一种灵活的架构，称为SPA-GCN，用于加速图形卷积网络（GCN），在图中的深度学习算法中的核心计算单元。该架构专门用于处理许多小图形，因为图表尺寸对设计考虑产生了重大影响。在这种情况下，我们使用SIMGNN是一种基于神经网络的图形匹配算法，作为展示我们架构的有效性的案例研究。实验结果表明，与多核CPU实施和GPU实施相比，SPA-GCN可以提供高速度，显示设计效率。

Graph neural networks (GNNs) have received great attention due to their success in various graph-related learning tasks. Several GNN frameworks have then been developed for fast and easy implementation of GNN models. Despite their popularity, they are not well documented, and their implementations and system performance have not been well understood. In particular, unlike the traditional GNNs that are trained based on the entire graph in a full-batch manner, recent GNNs have been developed with different graph sampling techniques for mini-batch training of GNNs on large graphs. While they improve the scalability, their training times still depend on the implementations in the frameworks as sampling and its associated operations can introduce non-negligible overhead and computational cost. In addition, it is unknown how much the frameworks are 'eco-friendly' from a green computing perspective. In this paper, we provide an in-depth study of two mainstream GNN frameworks along with three state-of-the-art GNNs to analyze their performance in terms of runtime and power/energy consumption. We conduct extensive benchmark experiments at several different levels and present detailed analysis results and observations, which could be helpful for further improvement and optimization.

随着图形神经网络（GNNS）在科学机器学习中的受欢迎程度的提高，他们的培训和推理效率变得越来越重要。此外，整个深度学习领域正在朝着更广泛和更深的网络趋向于越来越多的数据大小，以至于经常遇到硬件硬件瓶颈。新兴的专业硬件平台为这个问题提供了令人兴奋的解决方案。在本文中，我们系统地介绍并选择了与GNN有关的低级操作，以用于在Pytorch几何软件框架中实施的科学计算。然后，这些在NVIDIA A100 GPU上进行了严格的基准测试，以实现多种输入值组合，包括张量稀疏性。然后，我们为每个操作分析这些结果。在高水平上，我们得出结论，在NVIDIA系统上：（1）混淆瓶颈，例如记忆效率低下通常比单独的数据稀疏性占主导地位，（2）本地Pytorch操作通常比其Pytorch几何相等等等或更具竞争力。在低至中等水平的输入数据稀疏性下，以及（3）最新的GNN体系结构中心的许多操作几乎没有对稀疏性的优化。我们希望这些结果是那些在专门硬件上开发这些操作的人的基准，我们随后的分析有助于促进对这些操作的未来软件和基于硬件的优化，从而促进总体上可扩展的GNN性能。

Graph neural networks (GNNs) have been demonstrated to be a powerful algorithmic model in broad application fields for their effectiveness in learning over graphs. To scale GNN training up for large-scale and ever-growing graphs, the most promising solution is distributed training which distributes the workload of training across multiple computing nodes. However, the workflows, computational patterns, communication patterns, and optimization techniques of distributed GNN training remain preliminarily understood. In this paper, we provide a comprehensive survey of distributed GNN training by investigating various optimization techniques used in distributed GNN training. First, distributed GNN training is classified into several categories according to their workflows. In addition, their computational patterns and communication patterns, as well as the optimization techniques proposed by recent work are introduced. Second, the software frameworks and hardware platforms of distributed GNN training are also introduced for a deeper understanding. Third, distributed GNN training is compared with distributed training of deep neural networks, emphasizing the uniqueness of distributed GNN training. Finally, interesting issues and opportunities in this field are discussed.

图形神经网络（GNN）在处理图形结构数据的问题上表现出巨大的希望。 GNNS的独特点之一是它们的灵活性适应多个问题，这不仅导致广泛的适用性，而且在为特定问题找到最佳模型或加速技术时会带来重要的挑战。此类挑战的一个例子在于一个事实，即GNN模型或加速技术的准确性或有效性通常取决于基础图的结构。在本文中，为了解决图形依赖性加速的问题，我们提出了预后，这是一个数据驱动的模型，可以通过检查输入图来预测给定GNN模型在任意特征图上运行的GNN训练时间指标。这样的预测是基于先前使用多样化的合成图数据集经过离线训练的回归做出的。在实践中，我们的方法允许做出明智的决定，以用于特定问题的设计。在本文中，为特定用例定义并应用了构建预后的方法，其中有助于确定哪种图表更好。我们的结果表明，预后有助于在多种广泛使用的GNN模型（例如GCN，GIN，GAT或GRAPHSAGE）中随机选择图表的平均速度为1.22倍。

图形神经网络（GNNS）已成为处理机器学习任务的有效方法，它为构建推荐系统带来了一种新方法，其中可以将推荐任务作为用户 - 项目的链接预测问题提出， 。培训基于GNN的推荐系统（GNNRECSYS）在大图上会引起大型内存足迹，很容易超过典型服务器上的DRAM容量。现有的解决方案诉诸分布式子图培训，这是由于动态构建子图和各个子图的大量冗余的高成本而效率低下。新兴的Intel Optane持久记忆使一台机器以可承受的成本具有最多6 TB的存储器，从而使单机器Gnnrecsys训练可行，从而消除了分布式培训中的效率低下。与DRAM相比，将Optane用于Gnnrecsys的一个主要问题是Optane相对较低的带宽。由于其主要的计算内核稀疏且内存访问密集，因此这种限制可能对Gnnrecsys工作量的高性能特别有害。为了了解Optane是否适合Gnnrecsys培训，我们对Gnnrecsys工作负载进行了深入的表征和全面的基准测试研究。我们的基准测试结果表明，经过正确配置后，基于Optane的单机器GNNRECSYS训练优于大幅度的培训，尤其是在处理深度GNN模型时。我们分析了加速度的来源，提供有关如何为GNNRECSYS工作负载配置Optane的指导，并讨论进一步优化的机会。

图形神经网络（GNN）代表了在图形结构上运行的深度学习模型的新兴线路。由于其在许多与图形相关任务中实现的高精度，它变得越来越受欢迎。然而，在系统和建筑社区中，GNN在系统和建筑社区中不太了解，作为其对应物，例如多层的感知和卷积神经网络。这项工作试图向我们的社区介绍GNN。与仅呈现GCNS的特征的事先工作相比，我们的工作基于一般GNN描述框架覆盖GNN工作负载的大部分品种。通过构建两个广泛使用的库之上的模型，我们在有关通用和特定于应用程序的架构的推理阶段的特征在于GNN计算，希望我们的工作能够促进更多的系统和建筑研究GNN。

TensorFlow GNN（TF-GNN）是张量曲线的图形神经网络的可扩展库。它是从自下而上设计的，以支持当今信息生态系统中发生的丰富的异质图数据。Google的许多生产模型都使用TF-GNN，最近已作为开源项目发布。在本文中，我们描述了TF-GNN数据模型，其KERAS建模API以及相关功能，例如图形采样，分布式训练和加速器支持。

In the last few years, graph neural networks (GNNs) have become the standard toolkit for analyzing and learning from data on graphs. This emerging field has witnessed an extensive growth of promising techniques that have been applied with success to computer science, mathematics, biology, physics and chemistry. But for any successful field to become mainstream and reliable, benchmarks must be developed to quantify progress. This led us in March 2020 to release a benchmark framework that i) comprises of a diverse collection of mathematical and real-world graphs, ii) enables fair model comparison with the same parameter budget to identify key architectures, iii) has an open-source, easy-to-use and reproducible code infrastructure, and iv) is flexible for researchers to experiment with new theoretical ideas. As of December 2022, the GitHub repository has reached 2,000 stars and 380 forks, which demonstrates the utility of the proposed open-source framework through the wide usage by the GNN community. In this paper, we present an updated version of our benchmark with a concise presentation of the aforementioned framework characteristics, an additional medium-sized molecular dataset AQSOL, similar to the popular ZINC, but with a real-world measured chemical target, and discuss how this framework can be leveraged to explore new GNN designs and insights. As a proof of value of our benchmark, we study the case of graph positional encoding (PE) in GNNs, which was introduced with this benchmark and has since spurred interest of exploring more powerful PE for Transformers and GNNs in a robust experimental setting.

图形神经网络（GNNS）在学习从图形结构数据中展示了成功，其中包含欺诈检测，推荐和知识图形推理。然而，培训GNN有效地具有挑战性，因为：1）GPU存储器容量有限，对于大型数据集可能不足，而2）基于图形的数据结构导致不规则的数据访问模式。在这项工作中，我们提供了一种统计分析的方法，并确定了GNN培训前更频繁地访问的数据。我们的数据分层方法不仅利用输入图的结构，而且还从实际GNN训练过程中获得了洞察力，以实现更高的预测结果。通过我们的数据分层方法，我们还提供了一种新的数据放置和访问策略，以进一步最大限度地减少CPU-GPU通信开销。我们还考虑了多GPU GNN培训，我们也展示了我们在多GPU系统中的策略的有效性。评估结果表明，我们的工作将CPU-GPU流量降低了87-95％，并通过数亿节点和数十亿边缘的图表提高了现有解决方案的GNN训练速度。

Dynamic Graph Neural Networks (DGNNs) have been broadly applied in various real-life applications, such as link prediction and pandemic forecast, to capture both static structural information and temporal characteristics from dynamic graphs. Combining both time-dependent and -independent components, DGNNs manifest substantial parallel computation and data reuse potentials, but suffer from severe memory access inefficiency and data transfer overhead under the canonical one-graph-at-a-time training pattern. To tackle the challenges, we propose PiPAD, a $\underline{\textbf{Pi}}pelined$ and $\underline{\textbf{PA}}rallel$ $\underline{\textbf{D}}GNN$ training framework for the end-to-end performance optimization on GPUs. From both the algorithm and runtime level, PiPAD holistically reconstructs the overall training paradigm from the data organization to computation manner. Capable of processing multiple graph snapshots in parallel, PiPAD eliminates the unnecessary data transmission and alleviates memory access inefficiency to improve the overall performance. Our evaluation across various datasets shows PiPAD achieves $1.22\times$-$9.57\times$ speedup over the state-of-the-art DGNN frameworks on three representative models.

Graph convolutional neural networks (GCNs) have emerged as a key technology in various application domains where the input data is relational. A unique property of GCNs is that its two primary execution stages, aggregation and combination, exhibit drastically different dataflows. Consequently, prior GCN accelerators tackle this research space by casting the aggregation and combination stages as a series of sparse-dense matrix multiplication. However, prior work frequently suffers from inefficient data movements, leaving significant performance left on the table. We present GROW, a GCN accelerator based on Gustavson's algorithm to architect a row-wise product based sparse-dense GEMM accelerator. GROW co-designs the software/hardware that strikes a balance in locality and parallelism for GCNs, achieving significant energy-efficiency improvements vs. state-of-the-art GCN accelerators.

高吞吐量数据处理应用的高效硬件加速器设计，例如深度神经网络，是计算机架构设计中有挑战性的任务。在这方面，高级合成（HLS）作为快速原型设计的解决方案，从应用程序计算流程的行为描述开始。这种设计空间探索（DSE）旨在识别帕累托最佳的合成配置，其穷举搜索由于设计空间维度和合成过程的禁止计算成本而往往不可行。在该框架内，我们通过提出在文献中，有效和有效地解决了设计问题图形神经网络，该神经网络共同预测了合成的行为规范的加速性能和硬件成本给出了优化指令。考虑到性能和成本估计，学习模型可用于通过引导DSE来快速接近帕累托曲线。所提出的方法优于传统的HLS驱动DSE方法，通过考虑任意长度的计算机程序和输入的不变特性。我们提出了一种新颖的混合控制和数据流图表示，可以在不同硬件加速器的规格上培训图形神经网络;该方法自然地转移到解除数据处理应用程序。此外，我们表明我们的方法实现了与常用模拟器的预测准确性相当，而无需访问HLS编译器和目标FPGA的分析模型，同时是更快的数量级。最后，通过微调来自新目标域的少量样本，可以在未开发的配置空间中解放所学习的表示。

高级综合（HLS）释放了计算机架构师以非常低级的语言开发他们的设计，并需要准确指定如何在寄存器级别传输数据。在HLS的帮助下，硬件设计人员必须只描述设计的高级行为流程。尽管如此，它仍然可能需要数周才能开发高性能架构，主要是因为在更高的水平下有许多设计选择需要更多的时间来探索。它还需要几分钟才能从HLS工具上获得每个设计候选人的质量的反馈。在本文中，我们建议通过使用培训的图形神经网络（GNN）来建立HLS工具来解决这个问题，该工具被培训用于广泛的应用程序。实验结果表明，通过采用基于GNN的模型，我们能够以高精度估计毫秒的设计质量，这可以帮助我们非常快速地搜索解决方案空间。

Deep learning has revolutionized many machine learning tasks in recent years, ranging from image classification and video processing to speech recognition and natural language understanding. The data in these tasks are typically represented in the Euclidean space. However, there is an increasing number of applications where data are generated from non-Euclidean domains and are represented as graphs with complex relationships and interdependency between objects. The complexity of graph data has imposed significant challenges on existing machine learning algorithms. Recently, many studies on extending deep learning approaches for graph data have emerged. In this survey, we provide a comprehensive overview of graph neural networks (GNNs) in data mining and machine learning fields. We propose a new taxonomy to divide the state-of-the-art graph neural networks into four categories, namely recurrent graph neural networks, convolutional graph neural networks, graph autoencoders, and spatial-temporal graph neural networks. We further discuss the applications of graph neural networks across various domains and summarize the open source codes, benchmark data sets, and model evaluation of graph neural networks. Finally, we propose potential research directions in this rapidly growing field.

随着各个领域的深度学习的巨大成功，图形神经网络（GNNS）也成为图形分类的主要方法。通过全局读出操作，只会聚合所有节点（或节点群集）表示，现有的GNN分类器获得输入图的图级表示，并使用表示来预测其类标签。但是，这种全局聚合不考虑每个节点的结构信息，这导致全局结构的信息丢失。特别地，它通过对所有节点表示来强制执行分类器的相同权重参数来限制辨别力;在实践中，他们中的每一个都有助于不同于其结构语义的目标类别。在这项工作中，我们提出了结构性语义读数（SSREAD）来总结位置级节点表示，这允许为分类模拟特定位置的权重参数，以及有效地捕获与全局结构相关的图形语义。给定输入图，SSREAD旨在通过使用其节点与结构原型之间的语义对齐来识别结构上有意义的位置，该结构原型编码每个位置的原型特征。结构原型经过优化，以最小化所有训练图的对准成本，而其他GNN参数训练以预测类标签。我们的实验结果表明，SSREAD显着提高了GNN分类器的分类性能和可解释性，同时兼容各种聚合函数，GNN架构和学习框架。

稀疏的张量正在迅速成为现代深度学习工作负载的关键组成部分。但是，开发高性能的稀疏运营商可能很困难和乏味，现有的供应商库无法满足新运营商的不断升级要求。稀疏张量编译器简化了操作员的开发，但是对深度学习的有效稀疏编译仍然具有挑战性，因为单个稀疏格式无法最大程度地提高硬件效率，并且单次弹出编译器无法跟上最新的硬件和系统进步。我们表明，解决这两个挑战的关键是两种合成性。在本文中，我们提出了SparSetir，这是一种稀疏的张张汇编抽象，可为深度学习工作负载提供可合理的格式和可组合的转换。 Sparsetir在这些可组合组件上构建一个搜索空间，以进行性能调整。通过这些改进，SparSetir获得了单个操作员的GPU上的一致性能加速与供应商库：GNN操作员的1.1-3.3倍，稀疏变压器操作员的1.1-4.4x。 Sparsetir还以1.1-2.2倍的速度加速了端到端GNN，用于图形训练，而RGCN推断为0.9-26x。