The Research And Design Of VoD System Combined With CDN And P2P Technologies

Internet audio,video business is becoming an important network applications,but the server of the traditional unicast mode can easily become the system bottleneck. Content delivery network CDN and P2P networks technologies are currently the two mainstream providing audio and video services. The core of content distribution network is to publish media content quickly and efficiently to the nearest user service node, so the user can access the server rapidly, reducing the workload on the server and backbone networks and improving network utilization. P2P technology is based on " non-center" design and multicast mechanisms and weaken the role of the central server, allowing users to exchange information directly, sharing of resources rapidly, achieving a balanced workload.This article introduces the technique of CDN and P2P into VoD system, using thereliability,manageability of CDN technology, and the speed,scalability of P2Ptechnology, providing QoS assurance for the VoD system, and accommodating more users.The specific research content and work of this article include the following aspects:(1) The construction of video VoD system model combined with CDN and P2Ptechnologies. System model is consists of the resource server, edge server, tracker and the client, of which the resource server provides original audio and video resources, the edge servers provides users with quick starting playback of the initial data and the complementary streaming required for stable playback; center tracker provides resources index on the one hand, but also for user access control, ensuring the existing user' quality of service and accommodating more users; compared with the traditional model, the client also act as a temporary server.(2) The research and design of P2P resource node selection algorithm. Comparedwith the existing peer selection algorithm, the algorithm compositely considers the availability of resources and packet loss rate, available bandwidth, delay QoS targets. The specific of selection algorithm is completed in the tracker, the tracker is according to the requested resources segment, then find out a group of candidate resource nodes that obtain the requested resources segment, and according to the specific data weights and QoS targets for the best resource node. Here in order to prevent the tracker as a new bottleneck, we can use cluster technology.(3) System simulation and experiment. Based on the proposed system model,the simulation experiment results of access control and resource node selection algorithm show that, the system model is ensuring the existing nodes quality of service and accommodating more nodes.