In Search of the Performance- And Energy-Efficient CNN Accelerators
In this paper, starting from the algorithm, a performance- and energy-efficient 3D structure or shape of the Tensor Processing Engine (TPE) for CNN acceleration is systematically searched and evaluated. An optimal accelerator’s shape maximizes the number of concurrent MAC operations per clock cycle...
Main Authors: | Sedukhin, S. (Author), Tomioka, Y. (Author), Yamamoto, K. (Author) |
---|---|
Format: | Article |
Language: | English |
Published: |
Institute of Electronics Information Communication Engineers
2022
|
Subjects: | |
Online Access: | View Fulltext in Publisher |
Similar Items
-
Energy Efficiency Effects of Vectorization in Data Reuse Transformations for Many-Core Processors—A Case Study †
by: Abdullah Al Hasib, et al.
Published: (2017-02-01) -
Striping Input Feature Map Cache for Reducing off-chip Memory Traffic in CNN Accelerators
by: R. Struharik, et al.
Published: (2020-12-01) -
Iteration Time Prediction for CNN in Multi-GPU Platform: Modeling and Analysis
by: Ziqian Pei, et al.
Published: (2019-01-01) -
Parallel accelerator design for convolutional neural networks based on FPGA
by: Wang Ting, et al.
Published: (2021-02-01) -
SLID: Exploiting Spatial Locality in Input Data as a Computational Reuse Method for Efficient CNN
by: Fatmah Alantali, et al.
Published: (2021-01-01)