Machine Learning Driven Latency Optimization for Internet of Things Applications in Edge Computing

doi:10.12142/ZTECOM.202302007

ZTE Communications ›› 2023, Vol. 21 ›› Issue (2): 40-52.DOI: 10.12142/ZTECOM.202302007

• Special Topic • Previous Articles Next Articles

Machine Learning Driven Latency Optimization for Internet of Things Applications in Edge Computing

AWADA Uchechukwu¹, ZHANG Jiankang²(), CHEN Sheng^3,⁴, LI Shuangzhi¹, YANG Shouyi¹

^1.Zhengzhou University, Zhengzhou 450001, China
^2.Bournemouth University, Poole BH12 5BB, UK
^3.University of Southampton, Southampton SO17 1BJ, UK
^4.Ocean University of China, Qingdao 266100, China

Received:2023-03-11 Online:2023-06-13 Published:2023-06-13
About author:Uchechukwu AWADA is currently working toward a PhD degree at the School of Information Engineering, Zhengzhou University, China. His current research interests include edge computing, cloud computing, aerial computing, distributed systems, IoT, IoV and wireless communications. He is a student member of the ACM.|ZHANG Jiankang (jzhang3@bournemouth.ac.uk) is a senior lecturer at Bournemouth University, UK. Prior to joining Bournemouth University, he was a senior research fellow at the University of Southampton, UK. Dr. ZHANG was a lecturer from 2012 to 2013 and then an associate professor from 2013 to 2014 at Zhengzhou University, China. His research interests are in the areas of aeronautical communications, aeronautical networks, evolutionary algorithms and edge computing.|CHEN Sheng received his BE degree from the East China Petroleum Institute, China in 1982 and his PhD degree from City, University of London, UK in 1986, both in control engineering. In 2005, he was awarded the higher doctoral degree, Doctor of Sciences (DSc), from the University of Southampton, UK. From 1986 to 1999, He held research and academic appointments at the Universities of Sheffield, Edinburgh and Portsmouth, all in UK. Since 1999, he has been with the School of Electronics and Computer Science, the University of Southampton, where he holds the post of Professor in Intelligent Systems and Signal Processing. His research interests include adaptive signal processing, wireless communications, modeling and identification of nonlinear systems, neural network and machine learning, intelligent control system design, and evolutionary computation methods and optimization. He has published over 600 research papers. He has 18 500+ Web of Science citations with h-index of 59, and 36 700+ Google Scholar citations with h-index of 81. Dr. CHEN is a Fellow of the United Kingdom Royal Academy of Engineering, a Fellow of the Asia-Pacific Artificial Intelligence Association, and a Fellow of IET. He is one of the original ISI’s highly cited researchers in engineering (March 2004). He is named a 2023 Electronics and Electrical Engineering Leader in the UK by Research.com.|LI Shuangzhi received his BS and PhD degrees from the School of Information Engineering, Zhengzhou University, China in 2012 and 2018, respectively. From 2015 to 2017, he was a visiting student with the Department of Electrical and Computer Engineering, McMaster University, Canada. He is currently a lecturer with the School of Information Engineering, Zhengzhou University, China. His research interests include noncoherent space-time coding and ultra-reliable low-latency communications.|YANG Shouyi received his PhD degree from the Beijing Institute of Technology, China in 2002. He is currently a full professor with the School of Information Engineering, Zhengzhou University, China. He has authored or co-authored various articles in the field of signal processing and wireless communications. His current research interests include signal processing in communications systems, wireless communications, and cognitive radio.
Supported by:
the National Natural Science Foundation of China(61571401);61901416(part of the China Postdoctoral Science Foundation under Grant 2021TQ0304);the Innovative Talent Colleges and the University of Henan Province(18HASTIT021)

Abstract

Abstract:

Emerging Internet of Things (IoT) applications require faster execution time and response time to achieve optimal performance. However, most IoT devices have limited or no computing capability to achieve such stringent application requirements. To this end, computation offloading in edge computing has been used for IoT systems to achieve the desired performance. Nevertheless, randomly offloading applications to any available edge without considering their resource demands, inter-application dependencies and edge resource availability may eventually result in execution delay and performance degradation. We introduce Edge-IoT, a machine learning-enabled orchestration framework in this paper, which utilizes the states of edge resources and application resource requirements to facilitate a resource-aware offloading scheme for minimizing the average latency. We further propose a variant bin-packing optimization model that co-locates applications firmly on edge resources to fully utilize available resources. Extensive experiments show the effectiveness and resource efficiency of the proposed approach.

Key words: edge computing, execution time, IoT, machine learning, resource efficiency

AWADA Uchechukwu, ZHANG Jiankang, CHEN Sheng, LI Shuangzhi, YANG Shouyi. Machine Learning Driven Latency Optimization for Internet of Things Applications in Edge Computing[J]. ZTE Communications, 2023, 21(2): 40-52.

Figures/Tables 12

Figure 1 Directed acyclic graphs (DAG) of representative application

Figure 2 Application offloading strategies

Figure 3 An example architecture of Internet of Vehicles (IoV) multi-task offloading

Table 1 Notations

Notation	Description	Notation	Description
$E$	A set of edge deployments	$V$ , $V$	A vehicle, a set of vehicles
$T$	Individual application or task	$I i$	Container-instance or node in a cluster
$c, ? m$	CPU and memory resources	$I i c, ? m$	Resource capacity or availability of a node
$C$	A set of containerized applications	$C E d g e i c, ? m$	Resource capacity/availability in an edge
$d T c, ? m$	Application resource requirements	$U E d g e i c, ? m$	Resources used for execution
$E d g e i$	Individual edge deployment or cluster	$U E d g e i c$ , $U E d g e i m$	CPU, memory resource used for execution
$E d g e ?$	Closest edge deployment or cluster	$R U E d g e i c, ? m$	Actual resources usage of jobs
$R U E d g e i c$ , $R U E d g e i m$	Actual CPU, memory resources usage	$E s t$ , $E c p$	Application/task start, completion time
$E e x$	Application or task execution time	$U E d g e i ? c, ? m$	Cluster resource utilization
$U E d g e i ? c$ , $U E d g e i m$	Cluster CPU, memory resource utilization	$J$ , $J$	A job, a set of jobs

Table 1 Notations

Notation	Description	Notation	Description
$E$	A set of edge deployments	$V$ , $V$	A vehicle, a set of vehicles
$T$	Individual application or task	$I i$	Container-instance or node in a cluster
$c, ? m$	CPU and memory resources	$I i c, ? m$	Resource capacity or availability of a node
$C$	A set of containerized applications	$C E d g e i c, ? m$	Resource capacity/availability in an edge
$d T c, ? m$	Application resource requirements	$U E d g e i c, ? m$	Resources used for execution
$E d g e i$	Individual edge deployment or cluster	$U E d g e i c$ , $U E d g e i m$	CPU, memory resource used for execution
$E d g e ?$	Closest edge deployment or cluster	$R U E d g e i c, ? m$	Actual resources usage of jobs
$R U E d g e i c$ , $R U E d g e i m$	Actual CPU, memory resources usage	$E s t$ , $E c p$	Application/task start, completion time
$E e x$	Application or task execution time	$U E d g e i ? c, ? m$	Cluster resource utilization
$U E d g e i ? c$ , $U E d g e i m$	Cluster CPU, memory resource utilization	$J$ , $J$	A job, a set of jobs

Table 2 Edge deployments and their resource capacities

Edge Deployment	Edge Device	CPU Capacity	MemoryCapacity/GiB
Edge 1	Acer aiSage (x2)	12 Cores	4
Edge 2	AWS Snowcone (x10)	20 Cores	40
Edge 3	Huawei AR502H Series (x6)	24 Cores	12
Edge 4	HIVECELL (x6)	36 Cores	48
Edge 5	NVIDIA Jetson Xavier NX (x3)	36 Cores	24
Edge 6	INTELLIEDGE G700 (x5)	48 Cores	80

Table 3 Multi-job execution, where the actual resources consumed for multi-job execution dTc,?m??are taken from the original Alibaba data and the estimated resource demands dT?c,?m'are calculated by linear regression model

Multi-Job $J$	$C$	$T$	$d T ? ? c, ? m'$	$d T ? c, ? m'$	NAEE
$1$	$5$	$22$	$1 ? 195.24, ? 4.35$	$1 ? 135, ? 3.77$	$0.1, ? 0.15$
$2$	$7$	$29$	$1 ? 501.5, ? 5.81$	$1 ? 325, ? 4.23$	$0.13, ? 0.37$
$3$	$9$	$38$	$2 ? 011.55, ? 7.57$	$1 ? 820, ? 5.76$	$0.1, ? 0.3$
$4$	$12$	$52$	$2 ? 762.25, ? 10.4$	$2 ? 560, ? 8.2$	$0.1, ? 0.26$
$5$	$15$	$63$	$3 ? 369.68, ? 12.58$	$3185, ? 10.17$	$0.1, ? 0.23$

Multi-Job $J$	$C$	$T$	$d T ? ? c, ? m'$	$d T ? c, ? m'$	NAEE
$1$	$5$	$22$	$1 ? 195.24, ? 4.35$	$1 ? 135, ? 3.77$	$0.1, ? 0.15$
$2$	$7$	$29$	$1 ? 501.5, ? 5.81$	$1 ? 325, ? 4.23$	$0.13, ? 0.37$
$3$	$9$	$38$	$2 ? 011.55, ? 7.57$	$1 ? 820, ? 5.76$	$0.1, ? 0.3$
$4$	$12$	$52$	$2 ? 762.25, ? 10.4$	$2 ? 560, ? 8.2$	$0.1, ? 0.26$
$5$	$15$	$63$	$3 ? 369.68, ? 12.58$	$3185, ? 10.17$	$0.1, ? 0.23$

Figure 4 Tasks deployment ratio across the edge deployments

Figure 5 Average resource usage across the edge deployments

Figure 6 Average resource utilization across the edge deployments

Figure 7 Task scheduling time across edge deployments

Figure 8 Task execution time across edge deployments

Figure 9 Task response time across edge deployments

References 23

1	KHAN L U, YAQOOB I, TRAN N H, et al. Edge-computing-enabled smart cities: a comprehensive survey [J]. IEEE Internet of Things journal, 2020, 7(10): 10200–10232. DOI: 10.1109/JIOT.2020.2987070 DOI
2	AMIN S U, HOSSAIN M S. Edge intelligence and Internet of Things in healthcare: a survey [J]. IEEE access, 2020, 9: 45–59. DOI: 10.1109/ACCESS.2020.3045115 DOI
3	LIU Y J, WANG S G, ZHAO Q L, et al. Dependency-aware task scheduling in vehicular edge computing [J]. IEEE Internet of Things journal, 2020, 7(6): 4961–4971. DOI: 10.1109/JIOT.2020.2972041 DOI
4	SHEN Q Q, HU B J, XIA E J. Dependency-aware task offloading and service caching in vehicular edge computing [J]. IEEE transactions on vehicular technology, 2022, 71(12): 13182–13197. DOI: 10.1109/TVT.2022.3196544 DOI
5	REN H, LIU K, JIN F, et al. Dependency-aware task offloading via end-edge-cloud cooperation in heterogeneous vehicular networks [C]//25th International Conference on Intelligent Transportation Systems (ITSC). IEEE, 2022: 1420–1426. DOI: 10.1109/ITSC55140.2022.9922334 . DOI
6	LIU S S, LIU L K, TANG J, et al. Edge computing for autonomous driving: opportunities and challenges [J]. Proceedings of the IEEE, 2019, 107(8): 1697–1716. DOI: 10.1109/jproc.2019.2915983 DOI
7	MAHMUD R, TOOSI A N, RAMAMOHANARAO K, et al. Context-aware placement of industry 4.0 applications in fog computing environments [J]. IEEE transactions on industrial informatics, 2020, 16(11): 7004–7013. DOI: 10.1109/TII.2019.2952412 DOI
8	OTHMAN M M, EL-MOUSA A. Internet of Things & cloud computing Internet of Things as a service approach [C]//11th International Conference on Information and Communication Systems (ICICS). IEEE, 2020: 318–323. DOI: 10.1109/ICICS49469.2020.239503 DOI
9	REN J, ZHANG D Y, HE S W, et al. A survey on end-edge-cloud orchestrated network computing paradigms: transparent computing, mobile edge computing, fog computing, and cloudlet [J]. ACM computing surveys, 2020, 52(6): 1–36. DOI: 10.1145/3362031 DOI
10	HWANG J, NKENYEREYE L, SUNG N, et al. IoT service slicing and task offloading for edge computing [J]. IEEE Internet of Things journal, 2021, 8(14): 11526–11547. DOI: 10.1109/jiot.2021.3052498 DOI
11	ALMUTAIRI J, ALDOSSARY M. A novel approach for IoT tasks offloading in edge-cloud environments [J]. Journal of cloud computing, 2021, 10(1): 1–19. DOI: 10.1186/s13677-021-00243-9 DOI
12	AWADA U, ZHANG J K, CHEN S, et al. Air-to-air collaborative learning: a multi-task orchestration in federated aerial computing [C]//14th International Conference on Cloud Computing (CLOUD). IEEE, 2021: 671–680. DOI: 10.1109/CLOUD53861.2021.00086 DOI
13	AWADA U, ZHANG J K, CHEN S, et al. AirEdge: a dependency-aware multi-task orchestration in federated aerial computing [J]. IEEE transactions on vehicular technology, 2022, 71(1): 805–819. DOI: 10.1109/TVT.2021.3127011 DOI
14	TU Y F, DONG Z J, YANG H Z. Key Technologies and application of edge computing [J]. ZTE communications, 2017, 15(2): 26-34. DOI: 10.3969/j.issn.1673-5188.2017.02.004 DOI
15	LI X W, ZHAO L, YU K P, et al. A cooperative resource allocation model for IoT applications in mobile edge computing [J]. Computer communications, 2021, 173: 183–191. DOI: 10.1016/j.comcom.2021.04.005 DOI
16	LI J, LIANG W F, XU W Z, et al. Maximizing user service satisfaction for delay-sensitive IoT applications in edge computing [J]. IEEE transactions on parallel and distributed systems, 2022, 33(5): 1199–1212. DOI: 10.1109/TPDS.2021.3107137 DOI
17	ZHAN C, HU H, LIU Z, et al. Multi-UAV-enabled mobile-edge computing for time-constrained IoT applications [J]. IEEE Internet of Things journal, 2021, 8(20): 15553–15567. DOI: 10.1109/JIOT.2021.3073208 DOI
18	LI J, LIANG W F, XU W Z, et al. Service home identification of multiple-source IoT applications in edge computing [J]. IEEE transactions on services computing, 2023, 16(2): 1417–1430. DOI: 10.1109/TSC.2022.3176576 DOI
19	LIU J L, LIU C H, WANG B, et al. Optimized task allocation for IoT application in mobile-edge computing [J]. IEEE Internet of Things journal, 2022, 9(13): 10370–10381. DOI: 10.1109/JIOT.2021.3091599 DOI
20	HAN S N, LI X H, SUN C, et al. RecCac: Recommendation-empowered cooperative edge caching for internet of things [J]. ZTE communications, 2021, 19(2): 2–10. DOI: 10.12142/ZTECOM.202102002 DOI
21	LIU C H, LIU K, GUO S T, et al. Adaptive offloading for time-critical tasks in heterogeneous Internet of vehicles [J]. IEEE Internet of Things journal, 2020, 7(9): 7999–8011. DOI: 10.1109/JIOT.2020.2997720 DOI
22	RAMPERSAUD S, GROSU D. Sharing-aware online virtual machine packing in heterogeneous resource clouds [J]. IEEE transactions on parallel and distributed systems, 2017, 28(7): 2046–2059. DOI: 10.1109/TPDS.2016.2641937 DOI
23	HONG Z C, CHEN W H, HUANG H W, et al. Multi-hop cooperative computation offloading for industrial IoT-edge-cloud computing environments [J]. IEEE transactions on parallel and distributed systems, 2019, 30(12): 2759–2774. DOI: 10.1109/TPDS.2019.2926979 DOI

[1]	CHEN Jiajun, GAO Yin, LIU Zhuang, LI Dapeng. Future Vision on Artificial Intelligence Assisted Green Energy Efficiency Network [J]. ZTE Communications, 2023, 21(2): 34-39.
[2]	ZHAO Zipiao, ZHAO Yongli, YAN Boyuan, WANG Dajiang. Auxiliary Fault Location on Commercial Equipment Based on Supervised Machine Learning [J]. ZTE Communications, 2022, 20(S1): 7-15.
[3]	CAO Yinfeng, CAO Jiannong, WANG Yuqin, WANG Kaile, LIU Xun. Security in Edge Blockchains: Attacks and Countermeasures [J]. ZTE Communications, 2022, 20(4): 3-14.
[4]	NAN Yucen, FANG Minghao, ZOU Xiaojing, DOU Yutao, Albert Y. ZOMAYA. A Collaborative Medical Diagnosis System Without Sharing Patient Data [J]. ZTE Communications, 2022, 20(3): 3-16.
[5]	CUI Ziqi, WANG Gongpu, WANG Zhigang, AI Bo, XIAO Huahua. Symbiotic Radio Systems: Detection and Performance Analysis [J]. ZTE Communications, 2022, 20(3): 93-98.
[6]	HAN Suning, LI Xiuhua, SUN Chuan, WANG Xiaofei, LEUNG Victor C. M.. RecCac: Recommendation-Empowered Cooperative Edge Caching for Internet of Things [J]. ZTE Communications, 2021, 19(2): 2-10.
[7]	TAN Jie, SHA Xiubin, DAI Bo, LU Ting. Analysis of Industrial Internet of Things and Digital Twins [J]. ZTE Communications, 2021, 19(2): 53-60.
[8]	LIU Zhuang, GAO Yin, LI Dapeng, CHEN Jiajun, HAN Jiren. Enabling Energy Efficiency in 5G Network [J]. ZTE Communications, 2021, 19(1): 20-29.
[9]	LIN Xinhua, ZHANG Jing, LI Qiang. Cluster Head Selection Algorithm for UAV Assisted Clustered IoT Network Utilizing Blockchain [J]. ZTE Communications, 2021, 19(1): 30-38.
[10]	SHI Wenqi, SUN Yuxuan, HUANG Xiufeng, ZHOU Sheng, NIU Zhisheng. Scheduling Policies for Federated Learning in Wireless Networks: An Overview [J]. ZTE Communications, 2020, 18(2): 11-19.
[11]	WU Hequan. Ten Reflections on 5G [J]. ZTE Communications, 2020, 18(1): 1-4.
[12]	Julian AHRENS, Lia AHRENS, Hans D. SCHOTTEN. A Machine Learning Method for Prediction of Multipath Channels [J]. ZTE Communications, 2019, 17(4): 12-18.
[13]	LIU Jianwei, YUAN Yifei, HAN Jing. A Case Study on Intelligent Operation System for Wireless Networks [J]. ZTE Communications, 2019, 17(4): 19-26.
[14]	HAN Bin, Hans D. SCHOTTEN. Machine Learning for Network Slicing Resource Management:A Comprehensive Survey [J]. ZTE Communications, 2019, 17(4): 27-32.
[15]	XUE Songyan, LI Ang, WANG Jinfei, YI Na, MA Yi, Rahim TAFAZOLLI, Terence DODGSON. To Learn or Not to Learn:Deep Learning Assisted Wireless Modem Design [J]. ZTE Communications, 2019, 17(4): 3-11.

Machine Learning Driven Latency Optimization for Internet of Things Applications in Edge Computing

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 12

References 23

Related Articles 15

Recommended Articles

Metrics