Research And Implementation Of Universal Edge Resource Manager Under Two-Way Radio And Television Network

Nowadays, with the enhancement of people’s initiative access to information, the majority of TV users can no longer content to passively watch television programs, but hope to select a more different, more in line with personal preferences, more deepth information through the remote control. However, the current radio and television network architecture has been unable to meet this demand. The network is imperative to transform.The arrival of the tri-networks integration has brought an opportunity for the radio and television operators. They have stepped up to take this opportunity, the transformation of the two-way radio and television networks, originally only a one-way transmission line into a two-way line. Radio and television operators achieve the change from a single broadcast TV business into a comprehensive multi-service network through the transformation, and also the development of interactive on-demand and broadband data network based interactive two-way business.However, with the increase in the type of service, the scale of operations, the entire system also increases the demand for the various resources, of which the edge resources is the most important one. Now, in order to be able to use the same equipment to complete a variety of business, the operators use universal edge QAM. But various manufacturers have not published specific programs about how to manage univeral edge QAM for commercial confidentiality.This article is written under the above situation. By study the Comcast’s NGOD architecture and DOCSIS3.0M-CMTS architecture, this article designed to achieve a univeral edge resource manager for video-on-demand services, data services also the management and distribution on the universal edge device. In order to improve the resource utilization of the univeral edge device, this article studies several QAM resource allocation algorithms. At the same time, taking into account the growth in business and users, the concurrent performance of the univeral edge resource manager will become a bottleneck in the system. So this article studies the high concurrent server and analyzes several factors which affect the performance of server concurrency, such as frequent IO or frequent system calls.