Adapt-Flow: A Flexible DNN Accelerator Architecture for Heterogeneous Dataflow Implementation
Adapt-Flow: A Flexible DNN Accelerator Architecture for Heterogeneous Dataflow Implementation
复制标题
Adapt-Flow：用于异构数据流实现的灵活 DNN 加速器架构
DOI：
10.1145/3526241.3530311
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Louri, Ahmed
中科院分区：
文献类型：
--
作者：
Yang, Jiaqi;Zheng, Hao;Louri, Ahmed
深度神经网络（DNN）已被广泛应用于各种应用领域。DNN计算是内存和计算密集型的，需要过多的内存访问和大量的计算。为了有效地实现这些应用程序，已经提出了几种数据重用和并行开发策略，称为并行，。研究表明，许多DNN应用程序都受益于异构的低层策略，其中低层类型从一层到另一层发生变化。不幸的是，由于其有限的硬件灵活性，很少有现有的DNN架构可以同时容纳多个子网。在本文中，我们提出了一种灵活的DNN加速器架构，称为Adapt-Flow，它能够在运行时为每个DNN层支持多个并行选择。具体而言，所提出的Adapt-Flow架构包括（1）灵活的互连，（2）一个自适应流选择算法，和（3）自适应流映射技术。灵活的互连为不同业务流所需的各种业务模式提供动态支持。该算法针对DNN层选择最优的路由策略，从而大大提高了DNN的性能。而软流映射技术有效地将软流映射到柔性互连。仿真结果表明，与NVDLA、ShiDianNao和Eyeriss相比，Adapt-Flow的执行时间分别减少了46%、78%和26%，能耗分别减少了45%、80%和25%。
Deep neural networks (DNNs) have been widely applied to various application domains. DNN computation is memory and compute-intensive requiring excessive memory access and a large number of computations. To efficiently implement these applications, several data reuse and parallelism exploitation strategies, called dataflows, have been proposed. Studies have shown that many DNN applications benefit from a heterogeneous dataflow strategy where the dataflow type changes from layer to layer. Unfortunately, very few existing DNN architectures can simultaneously accommodate multiple dataflows due to their limited hardware flexibility. In this paper, we propose a flexible DNN accelerator architecture, called Adapt-Flow, which has the capability of supporting multiple dataflow selections for each DNN layer at runtime. Specifically, the proposed Adapt-Flow architecture consists of (1) a flexible interconnect, (2) a dataflow selection algorithm, and (3) a dataflow mapping technique. The flexible interconnect provides dynamic support for various traffic patterns required by different dataflows. The proposed dataflow selection algorithm selects the optimal dataflow strategy for a given DNN layer with the aim of much improved performance. And the dataflow mapping technique efficiently maps the dataflow amenable to the flexible interconnect. Simulation studies show that the proposed Adapt-Flow architecture reduces execution time by 46%, 78%, 26%, and energy consumption by 45%, 80%, 25% as compared to NVDLA, ShiDianNao, and Eyeriss respectively.
DOI：
10.1109/micro50266.2020.00062
发表时间：
2020-10
期刊：
2020 53rd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO)
影响因子：
--
作者：
Soroush Ghodrati;Byung Hoon Ahn;J. Kim;Sean Kinzer;B. Yatham;N. Alla;Hardik Sharma;Mohammad Alian;Eiman Ebrahimi;N. Kim;C. Young;H. Esmaeilzadeh
通讯作者：
Soroush Ghodrati;Byung Hoon Ahn;J. Kim;Sean Kinzer;B. Yatham;N. Alla;Hardik Sharma;Mohammad Alian;Eiman Ebrahimi;N. Kim;C. Young;H. Esmaeilzadeh