Partitioning multi-layer edge network for neural network collaborative computing

EURASIP Journal on Wireless Communications and Networking

Research

Tian-Feng Ren¹  Cheng-Bin Jiang¹  Ming-Tuo Zhou¹  Yong Chen¹  Qiang Li² 

[1]Jushri Technologies, Inc., Shanghai, China

[2]State Grid Information & Telecommunication Co., Ltd, Beijing, China

关键词: Neural network;    Edge computing;    Collaborative computing;    Genetic algorithm;   

DOI  :  10.1186/s13638-023-02284-x

 received in 2023-01-16, accepted in 2023-08-07,  发布年份 2023

来源: Springer

【 摘 要 】

There is a trend to deploy neural network on edge devices in recent years. While the mainstream of research often concerns with single edge device processing and edge-cloud two-layer neural network collaborative computing, in this paper, we propose partitioning multi-layer edge network for neural network collaborative computing. With the proposed method, sub-models of neural network are deployed on multi-layer edge devices along the communication path from end users to cloud. Firstly, we propose an optimal path selection method to form a neural network collaborative computing path with lowest communication overhead. Secondly, we establish a time-delay optimization mathematical model to evaluate the effects of different partitioning solutions. To find the optimal partition solution, an ordered elitist genetic algorithm (OEGA) is proposed. The experimental results show that, compared with traditional cloud computing, single-device edge computing and edge-cloud collaborative computing, the proposed multi-layer edge network collaborative computing has a smaller runtime delay with limited bandwidth resources, and because of the pipeline computing characteristics, the proposed method has a better response speed when processing large number of requests. Meanwhile, the OEGA algorithm has better performance than conventional methods, and the optimized partitioning method outperforms other methods like random and evenly partition.

【 授权许可】

CC BY   
© Springer Nature Switzerland AG 2023

【 图 表 】

Fig. 1

Fig. 2

Fig. 2

Fig. 1

Fig. 1

Fig. 3

Fig. 1

Fig. 4

Fig. 2

Fig. 3

Fig. 1

Fig. 2

Fig. 4

Fig. 5

Fig. 3

Fig. 2

12862_2023_2133_Article_IEq10.gif

Fig. 1

Fig. 3

Fig. 1

Fig. 2

Fig. 1

Fig. 4

40708_2023_198_Article_IEq16.gif

Fig. 3

Fig. 3

Fig. 4

12888_2023_5113_Article_IEq9.gif

Fig. 3

Fig. 3

Fig. 2

Fig. 4

Fig. 2

Fig. 3

Fig. 1

Fig. 2

Fig. 2

Fig. 1

Fig. 6

Fig. 1

Fig. 3

12888_2023_5113_Article_IEq8.gif

Fig. 3

12888_2023_5012_Article_IEq3.gif

Fig. 1

12888_2023_5012_Article_IEq2.gif

12888_2023_5113_Article_IEq7.gif

12888_2023_5113_Article_IEq6.gif

12888_2023_5113_Article_IEq4.gif

12888_2023_5113_Article_IEq2.gif

40517_2023_261_Article_IEq24.gif

40517_2023_261_Article_IEq45.gif

Fig. 3

40517_2023_261_Article_IEq43.gif

Fig. 2

Fig. 2

40517_2023_261_Article_IEq40.gif

Fig. 1

40517_2023_261_Article_IEq37.gif

40517_2023_261_Article_IEq36.gif

40517_2023_261_Article_IEq35.gif

40517_2023_261_Article_IEq34.gif

13690_2023_1154_Article_IEq14.gif

13690_2023_1154_Article_IEq13.gif

40517_2023_261_Article_IEq17.gif

13690_2023_1154_Article_IEq12.gif

13690_2023_1154_Article_IEq18.gif

13690_2023_1154_Article_IEq11.gif

Fig. 1

Fig. 1

Fig. 10

Fig. 5

Fig. 1

Fig. 6

Fig. 1

42490_2023_74_Article_IEq37.gif

42490_2023_74_Article_IEq18.gif

40517_2023_263_Article_IEq2.gif

Fig. 1

Fig. 3

Fig. 2

Fig. 1

41512_2023_153_Article_IEq93.gif

41512_2023_153_Article_IEq65.gif

13570_2023_282_Article_IEq6.gif

Fig. 3

Fig. 1

Fig. 1

Fig. 2