Small cell clustering for efficient distributed cloud computing

Femto-cloud is a novel networking architecture that joins femtocells networks and computation offloading to the cloud in a single framework. This allows to form server farms of femtocell access points providing cloud services. However, femtocells cannot offer the same computation and storage capacities as traditional cloud servers. In the femto-cloud platform, femtocells cooperate together through cluster formation. Effective cooperation between femtocells through clustering has to take into account many challenging limitations such as radio resources availability, base stations deployment scenarios, delay constraints and power consumption limitations. These parameters affect the choice of the femtocells cluster size and the computation load distribution. In this paper, we evaluate different strategies of clustering in the femto-cloud framework and show their effect on cluster characteristics in terms of size, latency, and power consumption.