Comparing Energy Efficiency of CPU, GPU and FPGA Implementations for Vision Kernels

被引:109
|
作者
Qasaimeh, Murad [1 ]
Denolf, Kristof [2 ]
Lo, Jack [2 ]
Vissers, Kees [2 ]
Zambreno, Joseph [1 ]
Jones, Phillip H. [1 ]
机构
[1] Iowa State Univ, Ames, IA 50011 USA
[2] Xilinx Res Labs, San Jose, CA USA
关键词
Embedded Vision; GPUs; FPGAs; CPUs; Energy Efficiency;
D O I
10.1109/icess.2019.8782524
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Developing high performance embedded vision applications requires balancing run-time performance with energy constraints. Given the mix of hardware accelerators that exist for embedded computer vision (e.g. multi-core CPUs, GPUs, and FPGAs), and their associated vendor optimized vision libraries, it becomes a challenge for developers to navigate this fragmented solution space. To aid with determining which embedded platform is most suitable for their application, we conduct a comprehensive benchmark of the run-time performance and energy efficiency of a wide range of vision kernels. We discuss rationales for why a given underlying hardware architecture innately performs well or poorly based on the characteristics of a range of vision kernel categories. Specifically, our study is performed for three commonly used HW accelerators for embedded vision applications: ARM57 CPU, Jetson TX2 GPU and ZCU102 FPGA, using their vendor optimized vision libraries: OpenCV, VisionWorks and xfOpenCV. Our results show that the GPU achieves an energy/frame reduction ratio of 1.1-3.2x compared to the others for simple kernels. While for more complicated kernels and complete vision pipelines, the FPGA outperforms the others with energy/frame reduction ratios of 1.2-22.3x. It is also observed that the FPGA performs increasingly better as a vision application's pipeline complexity grows.
引用
收藏
页数:8
相关论文
共 50 条
  • [1] Analyzing the Energy-Efficiency of Vision Kernels on Embedded CPU, GPU and FPGA Platforms
    Qasaimeh, Murad
    Zambreno, Joseph
    Jones, Phillip H.
    Denolf, Kristof
    Lo, Jack
    Vissers, Kees
    2019 27TH IEEE ANNUAL INTERNATIONAL SYMPOSIUM ON FIELD-PROGRAMMABLE CUSTOM COMPUTING MACHINES (FCCM), 2019, : 336 - 336
  • [2] FPGA, GPU, and CPU implementations of Jacobi algorithm for eigenanalysis
    Torun, Mustafa U.
    Yilmaz, Onur
    Akansu, Ali N.
    JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2016, 96 : 172 - 180
  • [3] CPU, GPU and FPGA Implementations of MALD: Ceramic Tile Surface Defects Detection Algorithm
    Matic, Tomislav
    Aleksi, Ivan
    Hocenski, Zeljko
    AUTOMATIKA, 2014, 55 (01) : 9 - 21
  • [4] Exploration of OpenCL 2D Convolution Kernels on Intel FPGA, CPU, and GPU Platforms
    Jin, Zheming
    Finkel, Hal
    2019 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2019, : 4460 - 4465
  • [5] Energy and Computing Assessment of Video Processing Kernels on CPU and FPGA platforms
    Mangrich, Fillipi
    Foes, Joao Gabriel Firta
    Correa, Guilherme
    Seidel, Ismael
    Grellert, Mateus
    2023 36TH SBC/SBMICRO/IEEE/ACM SYMPOSIUM ON INTEGRATED CIRCUITS AND SYSTEMS DESIGN, SBCCI, 2023, : 89 - 94
  • [6] BLAS Comparison on FPGA, CPU and GPU
    Kestur, Srinidhi
    Davis, John D.
    Williams, Oliver
    IEEE ANNUAL SYMPOSIUM ON VLSI (ISVLSI 2010), 2010, : 288 - 293
  • [7] Effectiveness of Fast Fourier Transform Implementations on GPU and CPU
    Puchala, Dariusz
    Stokfiszewski, Kamil
    Yatsymirskyy, Mykhaylo
    Szczepaniak, Bartlomiej
    2015 16TH INTERNATIONAL CONFERENCE ON COMPUTATIONAL PROBLEMS OF ELECTRICAL ENGINEERING (CPEE), 2015, : 162 - 164
  • [8] COMPARISON OF CPU AND GPU IMPLEMENTATIONS OF THE LATTICE BOLTZMANN METHOD
    McClure, James E.
    Prins, Jan F.
    Miller, Cass T.
    PROCEEDINGS OF THE XVIII INTERNATIONAL CONFERENCE ON COMPUTATIONAL METHODS IN WATER RESOURCES (CMWR 2010), 2010, : 1027 - 1034
  • [9] A Tutorial on the Implementations of Linear Image Filters in CPU and GPU
    Pardo, Alvaro
    COMPUTER SCIENCE (CACIC 2017), 2018, 790 : 111 - 121
  • [10] FPGA-Based On-Board Hyperspectral Imaging Compression: Benchmarking Performance and Energy Efficiency against GPU Implementations
    Caba, Julian
    Diaz, Maria
    Barba, Jesus
    Guerra, Raul
    de la Torre, Jose A.
    Lopez, Sebastian
    REMOTE SENSING, 2020, 12 (22) : 1 - 37