FPGA Based Accelerator for Neural Networks Computation with Flexible Pipelining

2021/12/28 by Qingyang Yi, Yi, Qingyang, Heming Sun +3
Computer Science · #FOS: Computer and information sciences #Hardware Architecture (cs.AR) #cs.AR

paper · pdf · doi:10.48550/arxiv.2112.15443

6 pages

arxiv created 2021/12/28 · arxiv updated 2022/01/03

Abstract

FPGA is appropriate for fix-point neural networks computing due to high power efficiency and configurability. However, its design must be intensively refined to achieve high performance using limited hardware resources. We present an FPGA-based neural networks accelerator and its optimization framework, which can achieve optimal efficiency for various CNN models and FPGA resources. Targeting high throughput, we adopt layer-wise pipeline architecture for higher DSP utilization. To get the optimal performance, a flexible algorithm to allocate balanced hardware resources to each layer is also proposed, supported by activation buffer design. Through our well-balanced implementation of four CNN models on ZC706, the DSP utilization and efficiency are over 90%. For VGG16 on ZC706, the proposed accelerator achieves the performance of 2.58x, 1.53x and 1.35x better than the referenced non-pipeline architecture [1], pipeline architecture [2] and [3], respectively.

Related

• FPGA-based Acceleration for Convolutional Neural Networks: A Comprehensive Review
• Maximizing CNN Accelerator Efficiency Through Resource Partitioning
• PipeCNN: An OpenCL-Based FPGA Accelerator for Large-Scale Convolution Neuron Networks
• A Survey of FPGA-Based Neural Network Accelerator
• ONNX-to-Hardware Design Flow for the Generation of Adaptive Neural-Network Accelerators on FPGAs