For more than a decade peer-to-peer networking technologies have met with huge success among Internet users, with peer-to-peer traffic already accounting for the largest part of the Internet's total traffic. Peer-to-peer networks have been extensively studied in recent years as an important supporting technology for state-of-the art applications, such as Live Video Streaming and Video-on-Demand. In this work the case of videoon-demand applications over BitTorrent-like unstructured peerto-peer networks is considered. As a rule, unstructured peerto-peer networks are pull-based, whereas push-based operation has been associated mostly with structured, tree-based p2p network topologies. While earlier studies maintained the pullbased nature of unstructured networks, recent ones have been investigating the combined use of pull and push operations, aiming at making more efficient use of the available bandwidth. Inspired by such hybrid approaches, this paper proposes a different use for push/pull protocols, uplink allocation based on user actions (i.e. random seek behavior) in video-on-demand applications. In a nutshell, this work proposes the employment of user-triggered push operations to prioritize content pre-fetching for the seemingly more popular parts for the users watching a video, thus increasing replication of such parts across the swarm.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.