Snowflake：一个模型无关的深度卷积神经网络加速器
Snowflake: A Model Agnostic Accelerator for Deep Convolutional Neural Networks
摘要：
深度卷积神经网络（CNN）是执行目标检测、分类、语义分割和自然语言处理任务的首选深度学习模型。CNN需要数十亿次操作来处理一帧。这种计算复杂性与卷积运算的固有并行性相结合，使CNN成为定制加速器的绝佳目标。然而，当针对不同的CNN层次结构和数据访问模式进行优化时，定制加速器很难达到接近100%的计算效率。在这项工作中，我们提出了Snowflake，这是一个可扩展且高效的加速器，对CNN工作负载不可知，并且旨在始终在接近峰值的硬件利用率下执行。Snowflake在现代CNN模型上的计算效率超过91%。在Xilinx Zynq XC 7Z 045 SoC上实现的Snowflake能够在AlexNet CNN模型上实现128 G-ops/s的峰值吞吐量和100帧每秒和120 G-ops/s的测量吞吐量，在GoogLeNet CNN模型上实现36帧每秒和116 G-ops/s的测量吞吐量，在ResNet-50 CNN模型上实现17帧每秒和122 G-ops/s的测量吞吐量。据我们所知，Snowflake是唯一一个能够在现代CNN上实现91%以上效率的实现系统，也是唯一一个将GoogLeNet和ResNet作为基准套件一部分的实现系统。
英文摘要：
Deep convolutional neural networks (CNNs) are the deep learning model of choice for performing object detection, classification, semantic segmentation and natural language processing tasks. CNNs require billions of operations to process a frame. This computational complexity, combined with the inherent parallelism of the convolution operation make CNNs an excellent target for custom accelerators. However, when optimizing for different CNN hierarchies and data access patterns, it is difficult for custom accelerators to achieve close to 100% computational efficiency. In this work, we present Snowflake, a scalable and efficient accelerator that is agnostic to CNN workloads, and was designed to always perform at near-peak hardware utilization. Snowflake is able to achieve a computational efficiency of over 91% on modern CNN models. Snowflake, implemented on a Xilinx Zynq XC7Z045 SoC is capable of achieving a peak throughput of 128G-ops/s and a measured throughput of 100 frames per second and 120 G-ops/s on the AlexNet CNN model, 36 frames per second and 116G- ops/s on the GoogLeNet CNN model and 17 frames per second and 122 G-ops/s on the ResNet-50 CNN model. To the best of our knowledge, Snowflake is the only implemented system capable of achieving over 91% efficiency on modern CNNs and the only implemented system with GoogLeNet and ResNet as part of the benchmark suite.
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。
