Convolutional Neural Network Accelerator with Reconfigurable Dataflow

Oh, Myungwoo; Lee, Chaeeun; Lee, Sang-Hun; Seo, Youngho; Kim, Sunwoo; Wang, Jooho; Sungchung, Chester

doi:10.1109/isocc.2018.8649988

Cited by 4 publications

(1 citation statement)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RS dataflow maps one row of filter data and one row of ifmap data to the corresponding processing unit, and the processing unit performs convolution operation to generate psum data. RS dataflow achieves multiplexing of filter data, multiplexing of ifmap data and minimization of psum data movement, which is the most efficient data flow at present [21,22].…”

Section: Related Workmentioning

confidence: 99%

An Efficient Dataflow Mapping Method for Convolutional Neural Networks

Liu

Zhang

et al. 2021

Neural Process Lett

View full text Add to dashboard Cite

Convolutional neural network (CNN) have been widely used in speech recognition, object detection and image recognition. In the process of inference, data access operations consume more energy than calculations. Therefore, optimizing the dataflow from external storage to the on-chip processing unit is an effective method to reduce the power consumption. The state-of-the-art row stationary dataflow maximizes the data reusing to reduce the number of data movement and thus reducing the power consumption of the system. However, it has the problems of low processing unit utilization and poor scalability. In this letter, we propose an enhance row stationary (ERS) dataflow. ERS dataflow maps the data of each channel of the three-dimensional filter and input feature map (ifmap) into a column of processing units by changing the mapping of data. The processing array operates multiple filters in parallel, which effectively improves the utilization of computing resources. Within each processing unit, one row of filter data and one row of ifmap data are processed at a time, and the filter row data and ifmap row data are reused in multiple convolution operations. In addition, a configurable sliding window model is proposed to solve the problem that the existing dataflow cannot handle the calculation array width smaller than the filter width. Simulation results show that for AlexNet and VGG16, the execution speed of ERS dataflow is improved by about 60% compared with row stationary dataflow, and the hardware resource utilization is improved by about 30%. For the MobileNet V1, the execution speed of ERS dataflow is improved by about 4% compared with the row stationary dataflow.

show abstract

Section: Related Workmentioning

confidence: 99%