Low Latency Deep Learning Inference Model for Distributed Intelligent IoT Edge Clusters
Edge computing is a new paradigm enabling intelligent applications for the Internet of Things (IoT) using mobile, low-cost IoT devices embedded with data analytics. Due to the resource limitations of Internet of Things devices, it is essential to use these resources optimally. Therefore, intelligenc...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9628063/ |
| _version_ | 1818405486912864256 |
|---|---|
| author | Soumyalatha Naveen Manjunath R. Kounte Mohammed Riyaz Ahmed |
| author_facet | Soumyalatha Naveen Manjunath R. Kounte Mohammed Riyaz Ahmed |
| author_sort | Soumyalatha Naveen |
| collection | DOAJ |
| description | Edge computing is a new paradigm enabling intelligent applications for the Internet of Things (IoT) using mobile, low-cost IoT devices embedded with data analytics. Due to the resource limitations of Internet of Things devices, it is essential to use these resources optimally. Therefore, intelligence needs to be applied through an efficient deep learning model to optimize resources like memory, power, and computational ability. In addition, intelligent edge computing is essential for real-time applications requiring end-to-end delay or response time within a few seconds. We propose decentralized heterogeneous edge clusters deployed with an optimized pre-trained yolov2 model. In our model, the weights have been pruned and then split into fused layers and distributed to edge devices for processing. Later the gateway device merges the partial results from each edge device to obtain the processed output. We deploy a convolutional neural network (CNN) on resource-constraint IoT devices to make them intelligent and realistic. Evaluation was done by deploying the proposed model on five IoT edge devices and a gateway device enabled with hardware accelerator. The evaluation of our proposed model shows significant improvement in terms of communication size and inference latency. Compared to DeepThings for <inline-formula> <tex-math notation="LaTeX">$5\times 5$ </tex-math></inline-formula> fused layer partitioning for five devices, our proposed model reduces communication size by ~ 14.4% and inference latency by ~16%. |
| first_indexed | 2024-12-14T08:56:49Z |
| format | Article |
| id | doaj.art-c30f70fec35d41a89ae807108d2717f5 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-14T08:56:49Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-c30f70fec35d41a89ae807108d2717f52022-12-21T23:08:54ZengIEEEIEEE Access2169-35362021-01-01916060716062110.1109/ACCESS.2021.31313969628063Low Latency Deep Learning Inference Model for Distributed Intelligent IoT Edge ClustersSoumyalatha Naveen0https://orcid.org/0000-0001-9552-3047Manjunath R. Kounte1https://orcid.org/0000-0002-2432-2552Mohammed Riyaz Ahmed2https://orcid.org/0000-0002-6061-6937School of Computer Science and Engineering, REVA University, Bengaluru, Karnataka, IndiaSchool of Electronics and Communication Engineering, REVA University, Bengaluru, Karnataka, IndiaSchool of Multidisciplinary Studies, REVA University, Bengaluru, Karnataka, IndiaEdge computing is a new paradigm enabling intelligent applications for the Internet of Things (IoT) using mobile, low-cost IoT devices embedded with data analytics. Due to the resource limitations of Internet of Things devices, it is essential to use these resources optimally. Therefore, intelligence needs to be applied through an efficient deep learning model to optimize resources like memory, power, and computational ability. In addition, intelligent edge computing is essential for real-time applications requiring end-to-end delay or response time within a few seconds. We propose decentralized heterogeneous edge clusters deployed with an optimized pre-trained yolov2 model. In our model, the weights have been pruned and then split into fused layers and distributed to edge devices for processing. Later the gateway device merges the partial results from each edge device to obtain the processed output. We deploy a convolutional neural network (CNN) on resource-constraint IoT devices to make them intelligent and realistic. Evaluation was done by deploying the proposed model on five IoT edge devices and a gateway device enabled with hardware accelerator. The evaluation of our proposed model shows significant improvement in terms of communication size and inference latency. Compared to DeepThings for <inline-formula> <tex-math notation="LaTeX">$5\times 5$ </tex-math></inline-formula> fused layer partitioning for five devices, our proposed model reduces communication size by ~ 14.4% and inference latency by ~16%.https://ieeexplore.ieee.org/document/9628063/Convolutional neural networkdeep learningdistributed intelligenceedge computingfog computingheterogeneous devices |
| spellingShingle | Soumyalatha Naveen Manjunath R. Kounte Mohammed Riyaz Ahmed Low Latency Deep Learning Inference Model for Distributed Intelligent IoT Edge Clusters IEEE Access Convolutional neural network deep learning distributed intelligence edge computing fog computing heterogeneous devices |
| title | Low Latency Deep Learning Inference Model for Distributed Intelligent IoT Edge Clusters |
| title_full | Low Latency Deep Learning Inference Model for Distributed Intelligent IoT Edge Clusters |
| title_fullStr | Low Latency Deep Learning Inference Model for Distributed Intelligent IoT Edge Clusters |
| title_full_unstemmed | Low Latency Deep Learning Inference Model for Distributed Intelligent IoT Edge Clusters |
| title_short | Low Latency Deep Learning Inference Model for Distributed Intelligent IoT Edge Clusters |
| title_sort | low latency deep learning inference model for distributed intelligent iot edge clusters |
| topic | Convolutional neural network deep learning distributed intelligence edge computing fog computing heterogeneous devices |
| url | https://ieeexplore.ieee.org/document/9628063/ |
| work_keys_str_mv | AT soumyalathanaveen lowlatencydeeplearninginferencemodelfordistributedintelligentiotedgeclusters AT manjunathrkounte lowlatencydeeplearninginferencemodelfordistributedintelligentiotedgeclusters AT mohammedriyazahmed lowlatencydeeplearninginferencemodelfordistributedintelligentiotedgeclusters |