Energy-Efficient Artificial Intelligence of Things With Intelligent Edge

被引：43

作者：

Zhu, Sha ^{[1
]}

Ota, Kaoru ^{[1
]}

Dong, Mianxiong ^{[1
]}

机构：

[1] Muroran Inst Technol, Dept Sci & Informat, Muroran, Hokkaido 0500071, Japan

来源：

IEEE INTERNET OF THINGS JOURNAL | 2022年 / 9卷 / 10期

关键词：

Artificial intelligence; Task analysis; Edge computing; Computational modeling; Cloud computing; Processor scheduling; Load modeling; Artificial Intelligence of Things (AIoT); energy efficiency; intelligent edge; IOT; SYSTEM;

D O I：

10.1109/JIOT.2022.3143722

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Artificial Intelligence of Things (AIoT) is an emerging area of future Internet of Things (IoT) to support intelligent IoT applications. In AIoT, intelligent edge computing technologies accelerate intelligent services' processing speed with much lower cost than simple cloud-aided IoT architecture. However, there is still a lack of resource strategy to optimize the energy efficiency of AIoT with intelligent edge computing. Therefore, in this article, we focus on the energy consumption of edge devices and cloud services in processing AIoT tasks and formulate the optimization problem in scheduling tasks in the edge and the cloud. Meanwhile, a novel online method is proposed to solve the optimization problem. We investigate the energy consumption of several typical intelligent edge devices and the cloud service in an intelligent edge computing testbed. Extensive simulation-based performance evaluation shows that the proposed method outperforms other strategies with lower energy consumption.

引用

页码：7525 / 7532

页数：8

共 34 条

[1]

Alvarez-Aparicio C., 2019, PROC IEEE INT C AUTO, P116

[2]

[Anonymous], 2016, P 14 ACM C EMB NETW, DOI DOI 10.1145/2994551.2994564

[3]

Arunkumar Akhil, 2017, ACM SIGARCH Computer Architecture News, V45, P320, DOI 10.1145/3140659.3080231

[4] DeepAD: A Generic Framework Based on Deep Learning for Time Series Anomaly Detection [J].

Buda, Teodora Sandra ;

Caglayan, Bora ;

Assem, Haytham .

ADVANCES IN KNOWLEDGE DISCOVERY AND DATA MINING, PAKDD 2018, PT I, 2018, 10937 :577-588

[5]

Chetlur Sharan, 2014, ARXIV

[6] Semisupervised Distributed Learning With Non-IID Data for AIoT Service Platform [J].

Chiu, Te-Chuan ;

Shih, Yuan-Yao ;

Pang, Ai-Chun ;

Wang, Chieh-Sheng ;

Weng, Wei ;

Chou, Chun-Ting .

IEEE INTERNET OF THINGS JOURNAL, 2020, 7 (10) :9266-9277

[7] GeePS: Scalable deep learning on distributed GPUs with a GPU-specialized parameter server [J].

Cui, Henggang ;

Zhang, Hao ;

Ganger, Gregory R. ;

Gibbons, Phillip B. ;

Xing, Eric P. .

PROCEEDINGS OF THE ELEVENTH EUROPEAN CONFERENCE ON COMPUTER SYSTEMS, (EUROSYS 2016), 2016,

[8] RMER: Reliable and Energy-Efficient Data Collection for Large-Scale Wireless Sensor Networks [J].

Dong, Mianxiong ;

Ota, Kaoru ;

Liu, Anfeng .

IEEE INTERNET OF THINGS JOURNAL, 2016, 3 (04) :511-519

[9] High-dimensional and large-scale anomaly detection using a linear one-class SVM with deep learning [J].

Erfani, Sarah M. ;

Rajasegarar, Sutharshan ;

Karunasekera, Shanika ;

Leckie, Christopher .

PATTERN RECOGNITION, 2016, 58 :121-134

[10]

Erol BA, 2018, STUD COMPUT INTELL, V777, P369, DOI 10.1007/978-3-319-89629-8_14

← 1 2 3 4 →