论文标题
用图神经网络的基准测试GPU和TPU性能
Benchmarking GPU and TPU Performance with Graph Neural Networks
论文作者
论文摘要
已经开发了许多人工智能（AI）设备来加速神经网络模型的训练和推理。最常见的是图形处理单元（GPU）和张量处理单元（TPU）。它们对密集的数据表示高度优化。但是，诸如图形之类的稀疏表示形式在包括科学在内的许多领域都普遍存在。因此，重要的是要在稀疏数据上表征可用的AI加速器的性能。这项工作分析并比较了GPU和TPU性能训练图形神经网络（GNN）开发的，以解决现实生活模式识别问题。表征作用于稀疏数据的新模型可能会有助于优化深度学习库和未来AI加速器的设计。
Many artificial intelligence (AI) devices have been developed to accelerate the training and inference of neural networks models. The most common ones are the Graphics Processing Unit (GPU) and Tensor Processing Unit (TPU). They are highly optimized for dense data representations. However, sparse representations such as graphs are prevalent in many domains, including science. It is therefore important to characterize the performance of available AI accelerators on sparse data. This work analyzes and compares the GPU and TPU performance training a Graph Neural Network (GNN) developed to solve a real-life pattern recognition problem. Characterizing the new class of models acting on sparse data may prove helpful in optimizing the design of deep learning libraries and future AI accelerators.
