Font Size: a A A

Research And Implementation On Data Caching Manage Policy For P2P Streaming Media System

Posted on:2015-11-25Degree:MasterType:Thesis
Country:ChinaCandidate:Z L WangFull Text:PDF
GTID:2298330422982048Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
With the development of Internet technology and the reinforcement of networkinfrastructure, the streaming media applications have rapidly developed, and are stillgrowing rapidly. On the one hand P2P streaming media technology plays more and moreimportant role in the development process; and on the other hand with the improving ofserver’s cost-effective, the role of the server in the streaming media system also has beenenhanced.We study caching strategies in Video-on-Demand streaming system, including thereplica placement strategies in streaming media server cluster and in the P2P network.Caching strategies is one of the important technologies related to the VoD streaming system.Through Optimizing replica placement strategies in the cluster, you can take advantage of acluster of storage space and network bandwidth, reduce the rate of denial of service,improve service capability of the system. P2P network client via the play-off in the localcache resources, not only to reduce server load, but also can improve the efficiency ofresource transfer of resources, reduce the pressure of the core network.Firstly, we study the replica placement strategies in streaming media server cluster,which include static replica placement policy (MMPacking algorithm) and dynamic replicaplacement strategies (placement strategy based on a copy of the threshold). We analyzetheir advantages and disadvantages. Then we study the traditional P2P streaming mediacache management mechanism.Secondly, we propose a cache management scheme considering both the cluster replicaand client cache in the P2P network. We consider the replicas in the server cluster, same asthe client cache, are special cache. We use different caching method for different resource,depending on resource popularity. And we use the threshold-based replication policy in theserver cluster and we use cache policy based on segment cache value. We give theirimplementation based on our project in the lab.Finally, the simulation experiments of the scheme are developed based on NS2. The results are analyzed, and show the proposed solution’s usefulness.
Keywords/Search Tags:Streaming Media System, Cache Algorithm, Replica ReplacementAlgorithm, Video-on-Demand, NS2
PDF Full Text Request
Related items