Chang-You Lin
ABSTRACT
Traditionally, deep learning acceleration mostly focuses on the trade-off between accuracy and training time but seldom addresses the deployment over hierarchical 5G networks to maximize the inference throughput. By contrast, computing offloading research emphasizes whether to offload the tasks to the cloud to reduce computing time and achieve a lower response time, and thus, the optimal deployment to maximize throughput has not been explored. In this paper, we explore Distributed Deep Neural Network Deployment Problem with Constrained Completion Time (TREND-WANT) to solve the deployment problem considering both response time and inference throughput. Due to the intractability of TREND-WANT, we first design a new algorithm, named Stage-Time-Aware Layer Deployment Algorithm (STEED), to maximize the throughput. Afterward, an extension termed STEED with Adaptable Completion Time (STEED-ADAPT) is developed to tailor the solution to achieve a lower responsible time. Simulation results manifest our algorithms outperform the traditional methods by at least 200%.
- 2019. Cisco Visual Networking Index: Global Mobile Data Traffic Forecast Update. White paper.Google Scholar
- 2019. Distributed Deep Neural Network Deployment for smart devices from the Edge to the Cloud (full). https://1drv.ms/b/s!AjOC7YQ-QPindxJGM4oiRmPE6ZUGoogle Scholar
- Jeffrey Dean et al. 2012. Large Scale Distributed Deep Networks. In NIPS.Google Scholar
- Forrest N. Iandola et al. 2016. FireCaffe: Near-Linear Acceleration of Deep Neural Network Training on Compute Clusters. In IEEE CVPR.Google Scholar
- Yiping Kang et al. 2017. Neurosurgeon: Collaborative Intelligence Between the Cloud and Mobile Edge. In ACM ASPLOS. Google ScholarDigital Library
- Andrew Zisserman Karen Simonyan. 2014. Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv: 1409.1556 (2014).Google Scholar
- J. Mao, Z. Qin, et al. 2017. AdaLearner: An adaptive distributed mobile learning system for neural networks. In IEEE/ACM ICCAD. Google ScholarDigital Library
- Cristina Marquez et al. 2018. How Should I Slice My Network? A Multi-Service Empirical Evaluation of Resource Sharing Efficiency. In ACM MOBICOM. Google ScholarDigital Library
- X. Ran, H. Chen, X. Zhu, Z. Liu, and J. Chen. 2018. Deep-Decision: A Mobile Deep Learning Framework for Edge Video Analytics. In IEEE INFOCOM.Google Scholar
- J. Redmon et al. 2016. You Only Look Once: Unified, Real-Time Object Detection. In CVPR.Google Scholar
- T. G. Rodrigues et al. 2017. Hybrid Method for Minimizing Service Delay in Edge Cloud Computing Through VM Migration and Transmission Power Control. IEEE Trans. Comput. 66 (2017), 810--819. Google ScholarDigital Library
- P. Rost et al. 2016. Mobile network architecture evolution toward 5G. IEEE Commun. Mag. 54 (2016), 84--91.Google ScholarCross Ref
- S. Teerapittayanon, B. McDanel, and H. T. Kung. 2017. Distributed Deep Neural Networks Over the Cloud, the Edge and End Devices. In IEEE ICDCS.Google Scholar
- J. Wang et al. 2018. Deep Learning towards Mobile Applications. In IEEE ICDCS.Google Scholar
- L. Wang et al. 2018. Service Entity Placement for Social Virtual Reality Applications in Edge Computing. In IEEE INFOCOM.Google Scholar
- S. Wang et al. 2018. When Edge Meets Learning: Adaptive Control for Resource-Constrained Distributed Machine Learning. In IEEE INFOCOM.Google Scholar
- J. Xu, L. Chen, and P. Zhou. 2018. Joint Service Caching and Task Offloading for Mobile Edge Computing in Dense Networks. In IEEE INFOCOM.Google Scholar
Index Terms
- Distributed Deep Neural Network Deployment for Smart Devices from the Edge to the Cloud
Recommendations
Cloud-based Enabling Mechanisms for Container Deployment and Migration at the Network Edge
SI: Evolution of IoT Networking Architectures papersIn recent years, a new trend of advanced applications with huge demands in terms of Quality of Service (QoS) is gaining ground. Even though Cloud computing provides mature management facilities with ubiquitous capabilities, novel requirements and ...
Towards the Decentralised Cloud: Survey on Approaches and Challenges for Mobile, Ad hoc, and Edge Computing
Cloud computing emerged as a centralised paradigm that made “infinite” computing resources available on demand. Nevertheless, the ever-increasing computing capacities present on smart connected things and devices calls for the decentralisation of Cloud ...
Security and Privacy Issues in Cloud, Fog and Edge Computing
AbstractThe advent of technologies like IoT and 5G brought a new computing paradigm called cloud computing into the world. Cloud computing has become the main platform for data warehousing and processing. However, storing data into the cloud has its own ...
Comments