Abstract
Recently, edge computing has become an emerging architectural pattern for mobile network operators (MNOs) to host services and perform related processing tasks at the base stations (BSs), which can reduce the volume of data transferred to the cloud via the network backhaul. The potential of edge computing can be fully exploited by coordinating the resources, e.g., computing, storage, and networking, at the BSs in a cooperative manner. In this paper, we explore this collaborative framework in the context of big data transferring. In particular, we aim to minimize the transferring cost by jointly optimizing the routes of the raw data, the services hosted in the BSs, and the amount of data to be processed. To this end, we formulate an optimization problem that captures the structures of both network flow problem and services placement problem. Given the NP-hardness of the formulated problem, a local search based heuristic is proposed, consisting of a diversification phase that generates multiple promising solutions and an intensification phase which is characterized by a hill climbing strategy. Numerical results are presented to demonstrate the effectiveness of the proposed approach.