Premium
Mobile agent‐based service migration in mobile edge computing
Author(s) -
Guo Yongan,
Jiang Chunlei,
Wu TinYu,
Wang Anzhi
Publication year - 2021
Publication title -
international journal of communication systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.344
H-Index - 49
eISSN - 1099-1131
pISSN - 1074-5351
DOI - 10.1002/dac.4699
Subject(s) - computer science , roaming , container (type theory) , mobile agent , flexibility (engineering) , distributed computing , mobile edge computing , service (business) , mobile technology , mobile computing , enhanced data rates for gsm evolution , mobile service , computer network , telecommunications , mechanical engineering , statistics , mathematics , economy , engineering , economics
Summary In order to improve the efficiency of service migration in mobile edge computing (MEC), and obtain higher quality of service and better user experience, we take mobile agent technology, instead of container, as the application component hosting technology. The advantages of mobile agent include small size, high flexibility, autonomy, and suitability for heterogeneous environments. For 5G communications, how to offer cross‐platform support and deal with frequent roaming of mobile nodes is our main concern. Therefore, this paper provides a service migration framework using mobile agent in MEC environment and performance evaluation. Experiments with tasks of different complexity show that in terms of efficiency of service migration, mobile agent technology obviously outperforms container technology. The main contributions of this paper include the following: (1) we solve the problem that too many auxiliary modules were used in the container architecture, (2) we use the agent container and the resource manager (RM) for performance evaluations to investigate the differences between mobile agent and container technology, and (3) we use the decision tree to confirm the execution cost of each node so as to analyze why the mobile agent responds to migration commands slowly.