Dynamic DNN Decomposition for Lossless Synergistic Inference

Uloženo v:
Podrobná bibliografie
Vydáno v:arXiv.org (Jan 15, 2021), p. n/a
Hlavní autor: Zhang, Beibei
Další autoři: Tian Xiang, Zhang, Hongxuan, Li, Te, Zhu, Shiqiang, Gu, Jianjun
Vydáno:
Cornell University Library, arXiv.org
Témata:
On-line přístup:Citation/Abstract
Full text outside of ProQuest
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

MARC

LEADER 00000nab a2200000uu 4500
001 2478673144
003 UK-CbPIL
022 |a 2331-8422 
035 |a 2478673144 
045 0 |b d20210115 
100 1 |a Zhang, Beibei 
245 1 |a Dynamic DNN Decomposition for Lossless Synergistic Inference 
260 |b Cornell University Library, arXiv.org  |c Jan 15, 2021 
513 |a Working Paper 
520 3 |a Deep neural networks (DNNs) sustain high performance in today's data processing applications. DNN inference is resource-intensive thus is difficult to fit into a mobile device. An alternative is to offload the DNN inference to a cloud server. However, such an approach requires heavy raw data transmission between the mobile device and the cloud server, which is not suitable for mission-critical and privacy-sensitive applications such as autopilot. To solve this problem, recent advances unleash DNN services using the edge computing paradigm. The existing approaches split a DNN into two parts and deploy the two partitions to computation nodes at two edge computing tiers. Nonetheless, these methods overlook collaborative device-edge-cloud computation resources. Besides, previous algorithms demand the whole DNN re-partitioning to adapt to computation resource changes and network dynamics. Moreover, for resource-demanding convolutional layers, prior works do not give a parallel processing strategy without loss of accuracy at the edge side. To tackle these issues, we propose D3, a dynamic DNN decomposition system for synergistic inference without precision loss. The proposed system introduces a heuristic algorithm named horizontal partition algorithm to split a DNN into three parts. The algorithm can partially adjust the partitions at run time according to processing time and network conditions. At the edge side, a vertical separation module separates feature maps into tiles that can be independently run on different edge nodes in parallel. Extensive quantitative evaluation of five popular DNNs illustrates that D3 outperforms the state-of-the-art counterparts up to 3.4 times in end-to-end DNN inference time and reduces backbone network communication overhead up to 3.68 times. 
653 |a Parallel processing 
653 |a Data processing 
653 |a Servers 
653 |a Electronic devices 
653 |a Artificial neural networks 
653 |a Partitions 
653 |a Cloud computing 
653 |a Edge computing 
653 |a Nodes 
653 |a Inference 
653 |a Feature maps 
653 |a Algorithms 
653 |a Data transmission 
653 |a Vertical separation 
653 |a Automatic pilots 
653 |a Heuristic methods 
653 |a Computer networks 
653 |a Run time (computers) 
653 |a Decomposition 
700 1 |a Tian Xiang 
700 1 |a Zhang, Hongxuan 
700 1 |a Li, Te 
700 1 |a Zhu, Shiqiang 
700 1 |a Gu, Jianjun 
773 0 |t arXiv.org  |g (Jan 15, 2021), p. n/a 
786 0 |d ProQuest  |t Engineering Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/2478673144/abstract/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch 
856 4 0 |3 Full text outside of ProQuest  |u http://arxiv.org/abs/2101.05952