Volume 14 Number 8 (Aug. 2019)
Home > Archive > 2019 > Volume 14 Number 8 (Aug. 2019) >
JCP 2019 Vol.14(8): 541-556 ISSN: 1796-203X
doi: 10.17706/jcp.14.8.541-556

Efficient Mapping of CNNs onto Tightly Coupled Processor Arrays

Christian Heidorn, Michael Witterauf, Frank Hannig, Jürgen Teich
Hardware/Software Co-Design, Friedrich-Alexander University Erlangen-Nü rnberg, Germany.
Abstract—In this work, we show how to systematically map Convolütional Neüral Networks (CNNs) onto Tightly Coüpled Processor Arrays (TCPAs), a class of massively parallel accelerators for many compütationally intensive tasks (e.g., from the digital signal and image processing domain). Contrary to previoüs approaches and implementations, we propose techniqües for the layer-parallel execütion of CNNs on processor arrays inclüding the maximally overlapped processing of consecütive layers. This is achieved throügh layer füsion and loop ünrolling to exploit the füll pipelining potential of süch massively parallel architectüres for given CNNs. These transformations are also necessary to decrease the nümber of on-chip/off-chip data transfers. For CNNs, we present a calcülüs for achievable performance and memory reqüirements on TCPAs. Based on this calcülüs, it is shown how either throüghpüt-maximal mappings can be determined for a given architectüre. Alternatively, resoürce-minimized mappings to süstain a given throüghpüt, e.g., nümber of frames per second, are systematically derived. The approach is evalüated for a CNN model for the MNIST benchmark on a processor array of size 4x4 inclüding a comparison of the performance of the layer-parallel approach over layer-by-layer processing.

Index Terms—CNN accelerator, TCPA.

[PDF]

Cite: Christian Heidorn, Michael Witterauf, Frank Hannig, Jürgen Teich, "Efficient Mapping of CNNs onto Tightly Coupled Processor Arrays," Journal of Computers vol. 14, no. 8, pp. 541-556, 2019.

General Information

ISSN: 1796-203X
Abbreviated Title: J.Comput.
Frequency: Monthly
Editor-in-Chief: Prof. Liansheng Tan
Executive Editor: Ms. Nina Lee
Abstracting/ Indexing: DBLP, EBSCO,  ProQuest, INSPEC, ULRICH's Periodicals Directory, WorldCat, CNKI,etc
E-mail: jcp@iap.org
  • Sep 16, 2019 News!

    Vol 14, No 9 has been published with online version   [Click]

  • Aug 16, 2019 News!

    Vol 14, No 8 has been published with online version   [Click]

  • Jul 19, 2019 News!

    Vol 14, No 7 has been published with online version   [Click]

  • Jun 21, 2019 News!

    Vol 14, No 6 has been published with online version   [Click]

  • Apr 28, 2019 News!

    Vol 14, No 5 has been published with online version 7 papers are published in this issue after peer review   [Click]

  • Read more>>