A Novel Deep Reinforcement Learning based service migration model for Mobile Edge Computing

被引:12
作者
Park, Sung Woon [1 ]
Boukerche, Azzedine [1 ]
Guan, Shichao [1 ]
机构
[1] Univ Ottawa, Sch Elect Engn & Comp Sci, Ottawa, ON, Canada
来源
PROCEEDINGS OF THE 2020 IEEE/ACM 24TH INTERNATIONAL SYMPOSIUM ON DISTRIBUTED SIMULATION AND REAL TIME APPLICATIONS (DS-RT) | 2020年
关键词
Mobile edge computing; service migration; deep reinforcement learning; energy consumption; migration cost; SCHEME;
D O I
10.1109/ds-rt50469.2020.9213536
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Cloud Computing has emerged as a foundation of smart environments by encapsulating and virtualizing the underlying design and implementation details. Concerning the inherent latency and deployment issues, Mobile Edge Computing seeks to migrate services in the vicinity of mobile users. However, the current migration-based studies lack the consideration of migration cost, transaction cost, and energy consumption on the system-level with discussion on the impact of personalized user mobility. In this paper, we implement an enhanced service migration model to address user proximity issues. We formalize the migration cost, transaction cost, energy consumption related to the migration process. We model the service migration issue as a complex optimization problem and adapt Deep Reinforcement Learning to approximate the optimal policy. We compare the performance of the proposed model with the recent Q-learning method and other baselines. The results demonstrate that the proposed model can estimate the optimal policy with complicated computation requirements.
引用
收藏
页码:84 / 91
页数:8
相关论文
共 28 条
[11]  
MacHen A., 2016, P ANN INT C MOB COMP
[12]   LIVE SERVICE MIGRATION IN MOBILE EDGE CLOUDS [J].
Machen, Andrew ;
Wang, Shiqiang ;
Leung, Kin K. ;
Ko, Bong Jun ;
Salonidis, Theodoros .
IEEE WIRELESS COMMUNICATIONS, 2018, 25 (01) :140-147
[13]  
Mnih V., 2013, PLAYING ATARI DEEP R
[14]   Human-level control through deep reinforcement learning [J].
Mnih, Volodymyr ;
Kavukcuoglu, Koray ;
Silver, David ;
Rusu, Andrei A. ;
Veness, Joel ;
Bellemare, Marc G. ;
Graves, Alex ;
Riedmiller, Martin ;
Fidjeland, Andreas K. ;
Ostrovski, Georg ;
Petersen, Stig ;
Beattie, Charles ;
Sadik, Amir ;
Antonoglou, Ioannis ;
King, Helen ;
Kumaran, Dharshan ;
Wierstra, Daan ;
Legg, Shane ;
Hassabis, Demis .
NATURE, 2015, 518 (7540) :529-533
[15]   Mobility Prediction Model-based Service Migration Procedure for Follow Me Cloud to support QoS and QoE [J].
Nadembega, Apollinaire ;
Hafid, Abdelhakim Senhaji ;
Brisebois, Ronald .
2016 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC), 2016,
[16]   Follow Me at the Edge: Mobility-Aware Dynamic Service Placement for Mobile Edge Computing [J].
Ouyang, Tao ;
Zhou, Zhi ;
Chen, Xu .
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, 2018, 36 (10) :2333-2345
[17]   Cloudlets Activation Scheme for Scalable Mobile Edge Computing with Transmission Power Control and Virtual Machine Migration [J].
Rodrigues, Tiago Gama ;
Suto, Katsuya ;
Nishiyama, Hiroki ;
Kato, Nei ;
Temma, Katsuhiro .
IEEE TRANSACTIONS ON COMPUTERS, 2018, 67 (09) :1287-1300
[18]   PRIMAL: PRofIt Maximization Avatar pLacement for Mobile Edge Computing [J].
Sun, Xiang ;
Ansari, Nirwan .
2016 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC), 2016,
[19]  
Sutton RS, 2018, ADAPT COMPUT MACH LE, P1
[20]  
Taleb T, 2013, IEEE GLOB COMM CONF, P1291, DOI 10.1109/GLOCOM.2013.6831252