Radio and Computing Resource Allocation for Minimizing Total Processing Completion Time in Mobile Edge Computing

Due to explosive growth in mobile applications and services with different requirements, the concept of mobile edge computing (MEC) has emerged. For MEC, a mobile user (MU) and a MEC server need to exchange tasks using limited radio resources. Furthermore, when multiple MUs possess tasks, the MEC se...

Full description

Saved in:
Bibliographic Details
Published in:IEEE access 2019, Vol.7, p.141119-141132
Main Authors: Kobayashi, Ryuji, Adachi, Koichi
Format: Article
Language:English
Subjects:
Algorithms
Applications programs
Completion time
Computational modeling
Edge computing
Mobile computing
Mobile edge computing
Mobile handsets
Optimization
Radio
Resource allocation
Resource management
Servers
Task analysis
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Due to explosive growth in mobile applications and services with different requirements, the concept of mobile edge computing (MEC) has emerged. For MEC, a mobile user (MU) and a MEC server need to exchange tasks using limited radio resources. Furthermore, when multiple MUs possess tasks, the MEC server has to handle multiple tasks simultaneously. Thus, the radio and computing resources need to be allocated to MUs by taking into account the wireless channel condition and the computing power of MUs and the MEC server. In this paper, a radio resource and computing resource allocation scheme is proposed to minimize the total processing completion time of all the tasks. Each task is assumed to be divided into local task and offload task. The local task is processed by each MU while the offload task is processed by a MEC server. We first formulate the optimization problem to minimize the total processing completion time of all tasks. To solve the formulated optimization problem, we propose a two-step radio and computing resources allocation scheme which iteratively performs bisection search method and Johnson's algorithm. The numerical results elucidate that the proposed scheme can reduce the total processing completion time by about 25% on average compared to the conventional schemes when multiple MUs have divisible tasks.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2019.2944184