arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.07894cs.ARcs.DCcs.LG
收藏

S2Engine:一种新的稀疏卷积神经网络系统结构

S2Engine: A Novel Systolic Architecture for Sparse Convolutional Neural Networks

Beihang University(北京航空航天大学)

Jianlei Yang, Wenzhi Fu, Xingzhou Cheng, Xucheng Ye, Pengcheng Dai, Weisheng Zhao

摘要:

卷积神经网络(CNN)在执行认知任务方面取得了巨大成功。然而,CNN的执行需要大量的计算资源,并产生大量的内存流量,这对计算系统设计提出了严峻的挑战。通过优化卷积的并行执行和数据重用,脉动架构在加速CNN计算方面表现出巨大的优势。然而,传统脉动架构中的规则内部数据传输路径阻止脉动架构完全利用由神经网络稀疏性引入的益处。部署细粒度稀疏现有的脉动架构是极大地阻碍了所产生的计算开销。在这项工作中,我们提出了S2引擎$-$一种新的脉动架构,可以充分利用CNN的稀疏性,最大限度地提高数据重用。S2Engine在内部传输压缩数据,并允许每个处理元件在卷积中动态地从压缩数据流中选择对齐的数据。与原始脉动阵列相比,S2引擎在速度和能效上分别实现了约$3.2\times$和约$3.0\times$的改进。

英文摘要:

Convolutional neural networks (CNNs) have achieved great success in performing cognitive tasks. However, execution of CNNs requires a large amount of computing resources and generates heavy memory traffic, which imposes a severe challenge on computing system design. Through optimizing parallel executions and data reuse in convolution, systolic architecture demonstrates great advantages in accelerating CNN computations. However, regular internal data transmission path in traditional systolic architecture prevents the systolic architecture from completely leveraging the benefits introduced by neural network sparsity. Deployment of fine-grained sparsity on the existing systolic architectures is greatly hindered by the incurred computational overheads. In this work, we propose S2Engine $-$ a novel systolic architecture that can fully exploit the sparsity in CNNs with maximized data reuse. S2Engine transmits compressed data internally and allows each processing element to dynamically select an aligned data from the compressed dataflow in convolution. Compared to the naive systolic array, S2Engine achieves about $3.2\times$ and about $3.0\times$ improvements on speed and energy efficiency, respectively.

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。

↑