Future applications such as driverless cars, industrial Internet and smart grids will demand high bandwidth, resilience, security and low latency communication at the same time. Those technical requirements will be met by the 5G communication system, which is not only focusing on the future air interface but also the deployment in the core network. The paradigm shift from agnostic store-and-forward towards intelligent networks paves the ground for the mobile edge cloud — a concept to not only place cloud computing in close proximity to the users or things, such as in the base stations or even in access points in the houses, but also move the cloud with the user as the device moves along the network. This would enable minimum latency services, which require advanced migration techniques that facilitate not just quick but very frequent migrations. In this paper, we study and compare the state of the art migration techniques offered by virtualization tools like docker and KVM and propose an application level migration protocol that eliminates the drawbacks of the former. We also present the implementation of the proposed protocol in a latency sensitive gaming application, where the server is migrated live during the gameplay between hosts, transparently to the users, as a proof of concept and study the handover in detail.
[1]
Ejaz Ahmed,et al.
A survey on mobile edge computing
,
2016,
2016 10th International Conference on Intelligent Systems and Control (ISCO).
[2]
Adam Wolisz,et al.
Providing application-level QoS in 3G/4G wireless systems: a comprehensive framework based on multirate CDMA
,
2002,
IEEE Wirel. Commun..
[3]
Andrew Warfield,et al.
Live migration of virtual machines
,
2005,
NSDI.
[4]
Simon M. Lucas,et al.
A UCT agent for Tron: Initial investigations
,
2010,
Proceedings of the 2010 IEEE Conference on Computational Intelligence and Games.
[5]
Frank H. P. Fitzek,et al.
Joint Design of Communication and Control for Connected Cars in 5G Communication Systems
,
2016,
2016 IEEE Globecom Workshops (GC Wkshps).
[6]
Rajeev Gandhi,et al.
The Case for Mobile Edge-Clouds
,
2013,
2013 IEEE 10th International Conference on Ubiquitous Intelligence and Computing and 2013 IEEE 10th International Conference on Autonomic and Trusted Computing.