The optimization game: Mobility and Edge computing in dense networks

The rise of mobile computing in the recent decades has given rise to a plethora of applications that are used everyday by millions of users. As these mobile applications have become more sophisticated, they have also become more resource intensive often requiring a fair share of computing resources and energy. Over the years, this requirement has been met with intelligent computation offloading to the edge and cloud with the devices themselves acting as portals to the data. However, as users move around while consuming services on their devices, the edge computing framework needs to move around based on the user’s present location in order to maintain the quality of service (QoS). While frequent migrations are required to maintain the QoS, the cost of such migrations add to the overall cost for the provider. As the movement of users across a service provider’s coverage area is unpredictable, we propose a computation offloading game to determine offloading decision process that is independent of any particular node and requires no a priori knowledge of the user’s future location. Our extensive analysis shows that our mechanism achieves efficient offloading performance and scales well with the system.