MASM: A Multiple-Algorithm Service Model for Energy-Delay Optimization in Edge Artificial Intelligence

Wenyu Zhang; Zhenjiang Zhang; Sherali Zeadally; Han Chieh Chao; Victor C.M. Leung

doi:10.1109/TII.2019.2897001

MASM: A Multiple-Algorithm Service Model for Energy-Delay Optimization in Edge Artificial Intelligence

Wenyu Zhang, Zhenjiang Zhang, Sherali Zeadally, Han Chieh Chao, Victor C.M. Leung

School of Information Science

Research output: Contribution to journal › Article › peer-review

84 Scopus citations

Abstract

Edge computing has emerged as a promising technique because of its advantages in providing low-latency computation offloading services for resource-limited mobile user devices and Internet of Things applications. Computationally intensive artificial intelligence (AI) tasks are well suited to be offloaded to the Cloudlet server, but there is a lack of energy-delay optimization models specifically designed for this edge AI scenario. In this paper, we propose a multiple algorithm service model (MASM) that provides heterogeneous algorithms with different computation complexities and required data sizes to fulfill the same task, and develop an optimization model that aims at reducing the energy and delay cost by optimizing the workload assignment weights and computing capacities of virtual machines, at the same time guaranteeing the quality of the results (QoRs). We propose a tide ebb algorithm to solve the MASM optimization model, and we prove its Parato optimality. Numerical results obtained demonstrate the effectiveness of our proposed method, and prove that the energy and delay costs can be significantly reduced by sacrificing the QoR of the offloaded AI tasks.

Original language	English
Article number	8632751
Pages (from-to)	4216-4224
Number of pages	9
Journal	IEEE Transactions on Industrial Informatics
Volume	15
Issue number	7
DOIs	https://doi.org/10.1109/TII.2019.2897001
State	Published - Jul 2019

Bibliographical note

Funding Information:
Manuscript received December 1, 2018; revised January 7, 2019; accepted January 27, 2019. Date of publication February 1, 2019; date of current version July 3, 2019. This work was supported in part by National Natural Science Foundation of China under Grant 61772064, and in part by the Academic Discipline, Post-Graduate Education Project of the Beijing Municipal Commission of Education. Paper no. TII-18-3219. (Corresponding author: Zhenjiang Zhang.) W. Zhang and Z. Zhang are with the School of Electronic and Information Engineering, Key Laboratory of Communication and Information Systems, Beijing Municipal Commission of Education, Beijing Jiaotong University, Beijing 100044, China (e-mail:, wenyuzhang@bjtu.edu.cn; zhjzhang1@bjtu.edu.cn).

Publisher Copyright:
© 2005-2012 IEEE.

Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.

Keywords

Artificial intelligence (AI)
edge computing
energy-delay optimization
resource management
workload assignment

ASJC Scopus subject areas

Control and Systems Engineering
Information Systems
Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/TII.2019.2897001

Cite this

@article{dee669233abe48d9bd5109d6c27c759a,

title = "MASM: A Multiple-Algorithm Service Model for Energy-Delay Optimization in Edge Artificial Intelligence",

abstract = "Edge computing has emerged as a promising technique because of its advantages in providing low-latency computation offloading services for resource-limited mobile user devices and Internet of Things applications. Computationally intensive artificial intelligence (AI) tasks are well suited to be offloaded to the Cloudlet server, but there is a lack of energy-delay optimization models specifically designed for this edge AI scenario. In this paper, we propose a multiple algorithm service model (MASM) that provides heterogeneous algorithms with different computation complexities and required data sizes to fulfill the same task, and develop an optimization model that aims at reducing the energy and delay cost by optimizing the workload assignment weights and computing capacities of virtual machines, at the same time guaranteeing the quality of the results (QoRs). We propose a tide ebb algorithm to solve the MASM optimization model, and we prove its Parato optimality. Numerical results obtained demonstrate the effectiveness of our proposed method, and prove that the energy and delay costs can be significantly reduced by sacrificing the QoR of the offloaded AI tasks.",

keywords = "Artificial intelligence (AI), edge computing, energy-delay optimization, resource management, workload assignment",

author = "Wenyu Zhang and Zhenjiang Zhang and Sherali Zeadally and Chao, {Han Chieh} and Leung, {Victor C.M.}",

note = "Funding Information: Manuscript received December 1, 2018; revised January 7, 2019; accepted January 27, 2019. Date of publication February 1, 2019; date of current version July 3, 2019. This work was supported in part by National Natural Science Foundation of China under Grant 61772064, and in part by the Academic Discipline, Post-Graduate Education Project of the Beijing Municipal Commission of Education. Paper no. TII-18-3219. (Corresponding author: Zhenjiang Zhang.) W. Zhang and Z. Zhang are with the School of Electronic and Information Engineering, Key Laboratory of Communication and Information Systems, Beijing Municipal Commission of Education, Beijing Jiaotong University, Beijing 100044, China (e-mail:, wenyuzhang@bjtu.edu.cn; zhjzhang1@bjtu.edu.cn). Publisher Copyright: {\textcopyright} 2005-2012 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.",

year = "2019",

month = jul,

doi = "10.1109/TII.2019.2897001",

language = "English",

volume = "15",

pages = "4216--4224",

number = "7",

}

TY - JOUR

T1 - MASM

T2 - A Multiple-Algorithm Service Model for Energy-Delay Optimization in Edge Artificial Intelligence

AU - Zhang, Wenyu

AU - Zhang, Zhenjiang

AU - Zeadally, Sherali

AU - Chao, Han Chieh

AU - Leung, Victor C.M.

N1 - Funding Information: Manuscript received December 1, 2018; revised January 7, 2019; accepted January 27, 2019. Date of publication February 1, 2019; date of current version July 3, 2019. This work was supported in part by National Natural Science Foundation of China under Grant 61772064, and in part by the Academic Discipline, Post-Graduate Education Project of the Beijing Municipal Commission of Education. Paper no. TII-18-3219. (Corresponding author: Zhenjiang Zhang.) W. Zhang and Z. Zhang are with the School of Electronic and Information Engineering, Key Laboratory of Communication and Information Systems, Beijing Municipal Commission of Education, Beijing Jiaotong University, Beijing 100044, China (e-mail:, wenyuzhang@bjtu.edu.cn; zhjzhang1@bjtu.edu.cn). Publisher Copyright: © 2005-2012 IEEE. Copyright: Copyright 2019 Elsevier B.V., All rights reserved.

PY - 2019/7

Y1 - 2019/7

N2 - Edge computing has emerged as a promising technique because of its advantages in providing low-latency computation offloading services for resource-limited mobile user devices and Internet of Things applications. Computationally intensive artificial intelligence (AI) tasks are well suited to be offloaded to the Cloudlet server, but there is a lack of energy-delay optimization models specifically designed for this edge AI scenario. In this paper, we propose a multiple algorithm service model (MASM) that provides heterogeneous algorithms with different computation complexities and required data sizes to fulfill the same task, and develop an optimization model that aims at reducing the energy and delay cost by optimizing the workload assignment weights and computing capacities of virtual machines, at the same time guaranteeing the quality of the results (QoRs). We propose a tide ebb algorithm to solve the MASM optimization model, and we prove its Parato optimality. Numerical results obtained demonstrate the effectiveness of our proposed method, and prove that the energy and delay costs can be significantly reduced by sacrificing the QoR of the offloaded AI tasks.

AB - Edge computing has emerged as a promising technique because of its advantages in providing low-latency computation offloading services for resource-limited mobile user devices and Internet of Things applications. Computationally intensive artificial intelligence (AI) tasks are well suited to be offloaded to the Cloudlet server, but there is a lack of energy-delay optimization models specifically designed for this edge AI scenario. In this paper, we propose a multiple algorithm service model (MASM) that provides heterogeneous algorithms with different computation complexities and required data sizes to fulfill the same task, and develop an optimization model that aims at reducing the energy and delay cost by optimizing the workload assignment weights and computing capacities of virtual machines, at the same time guaranteeing the quality of the results (QoRs). We propose a tide ebb algorithm to solve the MASM optimization model, and we prove its Parato optimality. Numerical results obtained demonstrate the effectiveness of our proposed method, and prove that the energy and delay costs can be significantly reduced by sacrificing the QoR of the offloaded AI tasks.

KW - Artificial intelligence (AI)

KW - edge computing

KW - energy-delay optimization

KW - resource management

KW - workload assignment

UR - http://www.scopus.com/inward/record.url?scp=85068615650&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068615650&partnerID=8YFLogxK

U2 - 10.1109/TII.2019.2897001

DO - 10.1109/TII.2019.2897001

M3 - Article

AN - SCOPUS:85068615650

SN - 1551-3203

VL - 15

SP - 4216

EP - 4224

JO - IEEE Transactions on Industrial Informatics

JF - IEEE Transactions on Industrial Informatics

IS - 7

M1 - 8632751

ER -

MASM: A Multiple-Algorithm Service Model for Energy-Delay Optimization in Edge Artificial Intelligence

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this