A bandwidth allocation scheme to meet flow requirements in mobile edge computing

Cloud computing has enabled users to enjoy various Internet services provided by data centers through a wide variety of access networks and communication terminals at anytime and anywhere. In cloud computing, the quality of latency-sensitive services depends on the distance between users and data centers. Therefore, mobile edge computing, which locates edge servers on access networks, has been proposed. It enables lower latency and higher-speed communication services than can be realized through computing solely through data centers. Various services in edge cloud computing, however, have different round-trip times (RTTs) and requirements. In this situation, an RTT-unfairness problem arises because such services commonly use TCP. In this study, we propose a bandwidth allocation scheme based on collectable information to meet the requirements of each flow in mobile edge computing; this is an extension of our previously reported scheme. We have confirmed the effectiveness of this approach through simulation evaluations.