Abstract

Abstract: A peer-to-peer non-linear media streaming systemneeds to schedule both on-demand and prefetch requests carefully so as toreduce the server load and ensure good user experience.A few change of thisreplication issues have been calculated recently with somewhat various result.Then they introduce  a merging requestscheduling model parameterized by the maximal number of peers that can be used to serve a single request . Thisarrangement is called Fair Sharing with Bounded Degree (FSBD). Based on thisunifying model, they  can analyze thevarious replication approach for different degree hop. Also it introduce asimple (primarily) distributed replication algorithm and show that thisalgorithm is able to modify itself to work well for different quality inarrangement. Based on this unifying model, we can compare the differentreplication strategies for different degree bounds and see how and whydifferent replication strategies are favored depending on the degree.We alsopropose a simple (primarily) distributed replication algorithm and show thatthis algorithm is able to adapt itself to work well for different degrees inscheduling.