NeuGraph: Parallel Deep Neural Network Computation on Large Graphs
NeuGraph: Parallel Deep Neural Network Computation on Large Graphs
复制标题
DOI：
--
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Lingxiao Ma;Zhi Yang;Youshan Miao;Jilong Xue;Ming Wu;Lidong Zhou;Yafei Dai
中科院分区：
文献类型：
--
作者：
Lingxiao Ma;Zhi Yang;Youshan Miao;Jilong Xue;Ming Wu;Lidong Zhou;Yafei Dai
最近的深度学习模型已经超越了低维规则网格（如图像、视频和语音），转向了高维图结构数据（如社交网络、电子商务用户项图和知识图）。这种演变导致了基于图的大型神经网络模型，这些模型超出了现有深度学习框架或图计算系统的设计目标。我们提出了NeuGraph，一个新的框架，它桥接了图和卷积模型，以支持图上的高效和可扩展的并行神经网络计算。NeuGraph将图计算优化引入基于区块链的深度学习框架中的数据分区、调度和并行管理。我们的评估表明，在可以容纳在单个GPU中的小图形上，NeuGraph的性能远远优于最先进的实现，同时可以扩展到现有框架都无法直接使用GPU处理的大型真实世界图形。（请继续关注进一步的更新。
Recent deep learning models have moved beyond low dimensional regular grids such as image, video, and speech, to high-dimensional graph-structured data, such as social networks, e-commerce user-item graphs, and knowledge graphs. This evolution has led to large graph-based neural network models that go beyond what existing deep learning frameworks or graph computing systems are designed for. We present NeuGraph, a new framework that bridges the graph and dataflow models to support efficient and scalable parallel neural network computation on graphs. NeuGraph introduces graph computation optimizations into the management of data partitioning, scheduling, and parallelism in dataflow-based deep learning frameworks. Our evaluation shows that, on small graphs that can fit in a single GPU, NeuGraph outperforms state-of-the-art implementations by a significant margin, while scaling to large real-world graphs that none of the existing frameworks can handle directly with GPUs. (Please stay tuned for further updates.)