A configurable SIMD architecture with explicit datapath for intelligent learning

Y. He, M.C.J. Peemen, L.J.W. Waeijen, E. Diken, M. Fiumara, G. Rauwerda, H. Corporaal, T. Geng

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

2 Citations (Scopus)
1 Downloads (Pure)

Abstract

The use of a wide Single-Instruction-Multiple-Data (SIMD) architecture is a promising approach to build energy-efficient high performance embedded processors. In this paper, based on our design framework for low-power SIMD processors, we propose a multiply-accumulate (MAC) unit with variable number of accumulator registers. The proposed MAC unit exploits both the merits of merged operation and register tiling. A Convolutional Neural Network (CNN) is a popular learning based algorithm due to its flexibility and high accuracy. However, a CNN-based application is often computationally intensive as it applies convolution operations extensively on a large data set. In this work, a CNN-based intelligent learning application is analyzed and mapped in the context of SIMD architectures. Experimental results show that the proposed architecture is efficient. In a 64-PE instance, the proposed SIMD processor with MAC4reg achieves an effective performance of 63.2 GOPS. Compared to the two baseline SIMD processors without MAC4reg, the proposed design brings 54.0% and 32.1% reduction in execution time, and 20.5% and 35.1% reduction in energy consumption respectively.

Original languageEnglish
Title of host publicationProceedings - 2016 16th International Conference on Embedded Computer Systems
Subtitle of host publicationArchitectures, Modeling and Simulation, SAMOS 2016
PublisherInstitute of Electrical and Electronics Engineers
Pages156-163
Number of pages8
ISBN (Electronic)9781509030767
DOIs
Publication statusPublished - 13 Jan 2017
Event16th International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation (SAMOS 2016), 18-21 July 2016, Samos, Greece - Samos, Greece
Duration: 18 Jul 201621 Jul 2016
http://www.wikicfp.com/cfp/servlet/event.showcfp?eventid=50851&copyownerid=84389

Conference

Conference16th International Conference on Embedded Computer Systems: Architectures, Modeling and Simulation (SAMOS 2016), 18-21 July 2016, Samos, Greece
Abbreviated titleSAMOS2016
CountryGreece
CitySamos
Period18/07/1621/07/16
Internet address

Keywords

  • CNN
  • Configurable
  • Explicit Datapath
  • Low Power
  • MAC
  • Wide SIMD

Fingerprint Dive into the research topics of 'A configurable SIMD architecture with explicit datapath for intelligent learning'. Together they form a unique fingerprint.

Cite this