MeDNN: A distributed mobile system with enhanced partition and deployment for large-scale DNNs

Published

Conference Paper

© 2017 IEEE. Deep Neural Networks (DNNs) are pervasively used in a significant number of applications and platforms. To enhance the execution efficiency of large-scale DNNs, previous attempts focus mainly on client-server paradigms, relying on powerful external infrastructure, or model compression, with complicated pre-processing phases. Though effective, these methods overlook the optimization of DNNs on distributed mobile devices. In this work, we design and implement MeDNN, a local distributed mobile computing system with enhanced partitioning and deployment tailored for large-scale DNNs. In MeDNN, we first propose Greedy Two Dimensional Partition (GTDP), which can adaptively partition DNN models onto several mobile devices w.r.t. individual resource constraints. We also propose Structured Model Compact Deployment (SMCD), a mobile-friendly compression scheme which utilizes a structured sparsity pruning technique to further accelerate DNN execution. Experimental results show that, GTDP can accelerate the original DNN execution time by 1.86-2.44x with 2-4 worker nodes. By utilizing SMCD, 26.5% of additional computing time and 14.2% of extra communication time are saved, on average, with negligible effect on the model accuracy.

Full Text

Duke Authors

Cited Authors

  • Mao, J; Yang, Z; Wen, W; Wu, C; Song, L; Nixon, KW; Chen, X; Li, H; Chen, Y

Published Date

  • December 13, 2017

Published In

Volume / Issue

  • 2017-November /

Start / End Page

  • 751 - 756

International Standard Serial Number (ISSN)

  • 1092-3152

International Standard Book Number 13 (ISBN-13)

  • 9781538630938

Digital Object Identifier (DOI)

  • 10.1109/ICCAD.2017.8203852

Citation Source

  • Scopus