Research On Design And Application Of Mobi-Le Edge Computing Network Based On SDN

With the rapid development of mobile Internet,mobile data traffic is showing an explosive growth,which has brought great pressure to the the core network;at the same time,the emergence of delay-sensitive services,such as augmented reality and HD video distribution,have raised higher performance requirements for the network.Traditional mobile cloud computing technology,of which computing mode together with data stream processing in remote cloud platform are no longer fit for the current situation.Mobile Edge Computing(MEC)is a network architecture that implements cloud computing capabilities and IT service environments at the edge of mobile networks,by deploying third-party applications,some core business processing and dispatching capabilities to the edge of the network near the user side to provide a network environment with low delay and high bandwidth.Due to the need to deploy cloud-like resources and associated IT services on the edge of the mobile network,and in order to ensure effective coordination between distributed resources,it will inevitably brings new challenges in network management.Therefore,in order to effectively integrate the edge network and cloud computing,and realize the flexible networking,a control mechanism with the ability of orchestrating distributed environment needs to be introduced in the network.The advantages of Software Defined Networking meet these requirements,the latest form of which has the inherent ability to mitigate barriers that prevent edge computing from reaching its full potential.All data flow control,business choreography,and other management tasks are done by the central SDN controller that is transparent to the end user.In view of the above discussion,this article mainly studies the combination of SDN and MEC technologies and their applications.The details are as follows:(1)Researching deeply on MEC,SDN related technical background as well as existing SDN-based edge computing solutions in academia,which provides a theoretical basis for the study of the combination of SDN and MEC.(2)Combined with the needs of future networks,this paper proposes a mobile edge computing network model based on SDN.The model uses virtualization technology to abstract resources and migrate cloud computing to the edge network;integrated scalable SDN controller,NFV orchestration mechanism for resource management and orchestration,to achieve collaboration between the edge network and the edge of the cloud,and make the network scalable;the northbound programmable application interface provides the ability to call the underlying network resources for the upper application.(3)According to the networking model,a low latency networking scheme is designed.In this scheme,Using the open source cloud management platform of OpenStack to implement the management of edge cloud platform and the deployment of service;the core function module of the SDN controller OpenDaylight is extended to realize the dynamic construction of low time delay network according to the user's needs,and support the implementation of the networking scheme;Using Extend ODL and the NFV orchestrator realize the integration of the edge network and the edge cloud to manage and arrange the global resources.(4)A set of network testing platform supporting multiple access is built,and a series of experiments are carried out to test and evaluate the delay performance of test platform.Developing high-definition video distribution applications and deploy them to the test platform and remote Alibaba cloud platform respectively,under the coordination of SDN controller,the video playback in different platforms is quantitatively analyzed.Experiments show that compared with the traditional mobile cloud computing,the networking scheme proposed in this paper has better performance advantages and can effectively carry delay-sensitive services.