Research project The goal of this project is to design new resource allocation methods to improve and unify the management of mobile edge clouds.
Mobile Edge Cloud (MEC) is an emerging paradigm combining computing, communication and storage capacity at the edge of the network with data-centers, in an elastic infrastructure. MECs have potential to host both throughput-oriented applications, requiring large network bandwidth and computing capacity, as well as latency-oriented applications, requiring low-latency communication and computation in end-user proximity. Most commonly, the computing capacity in edge locations and the wireless access are managed separately. However, delivering the full potential of MECs requires that edge locations and wireless networks be managed in concert.
The goal of this project is to design new resource allocation methods to improve and unify the management of MECs. Devising such methods is challenging, since delivering high quality service to the end-user depends on several players with partially conflicting goals, such as application providers, network operators, and infrastructure owners, each controlling a piece of the system. Thus, the problem needs to be solved at the two levels of mechanism and interaction.
First, assuming rational players, the issue is to come up with a mechanism that incentivizes truthfulness and maximizes social welfare. Game theory will be used to analyze the optimal strategy of each player, thus verifying that the goals of the devised mechanisms are met.
Second, for practical applicability an interaction protocol needs to be devised. This entails detailing what information should be transmitted between the players and in what order. The interaction protocol, which of course reinforces the mechanism, will be validated using simulations.
The project aims to (i) develop mathematical models and algorithms to efficiently solve the network and computing resource allocation for MEC to support emerging mobile applications and (ii) realize and implement the models and algorithms into software to demonstrate their feasibility and practicability. Therefore, the deliverables of the project will be both publications in top-tier journals and conferences as well as a workable prototype.