P2P caching

Abstract: In this paper, we present SmoothCache, a peer-to-peer live video streaming (P2PLS) system. The novelty of SmoothCache is threefold: i ) It is the first P2PLS system that is built to support the relatively-new approach of using HTTP as the transport protocol for live content, ii ) The system supports both single and multi-bitrate streaming modes of operation, and iii ) In Smoothcache, we make use of recent advances in application-layer dynamic congestion control to manage priorities of transfers according to their urgency. We start by explaining why the HTTP live streaming semantics render many of the existing assumptions used in P2PLS protocols obsolete. Afterwards, we present our design starting with a baseline P2P caching model. We, then, show a number of optimizations related to aspects such as neighborhood management, uploader selection and proactive caching. Finally, we present our evaluation conducted on a real yet instrumented test network. Our results show that we can achieve substantial traffic savings on the source of the stream without major degradation in user experience.

Abstract: The invention relates to a method for improving P2P flow caching efficiency, which is applicable in Internet content cache field. The method for improving P2P flow caching efficiency includes a P2P detection and control server, a P2P caching proxy server, a tracking server; wherein the P2P detection and control server is used to monitor and control P2P downloading and to reset the direction of P2P downloading; the P2P caching proxy server caches the P2P files and provides P2P downloading service for a user; and the tracking server guides the user to be connected with a Peer and a cache server within the network.

Abstract: Peer-to-peer (P2P) cache techniques for information-centric networks (ICN) provide the local data resources services for mobile users by storing the popular data contents in their peer mobile users' cache stations, so that a mobile user can obtain these data contents directly from its peer mobile users instead of the remote service provider. Retrieving the requested data content from the nearby storage locations (i.e., cache stations), the P2P caching technique improves the network performance by reducing the duplicate data transmission traffic. To overcome the difficulties of minimizing the data transmission delay for the P2P caching system, we divide all mobile users in a wireless cell as several P2P caching groups, where a mobile user can obtain its requested data content from a peer mobile user in the same group, or from a peer mobile user in another group. We propose the Markov decision process (MDP) based stochastic optimization scheme in this cache system with dynamic number of peers to derive the optimal lifespan for cached data contents in cache stations. The objective of our proposed stochastic optimization framework is to efficiently minimize the cost function, which jointly measures the cache size/cost and the data content transmission delay, over all P2P caching groups in the wireless cell. To achieve this goal, we apply the finite-state Markov chain to characterize the number of mobile users in each caching group and dynamically adapt the cached data lifespan in cache stations. We also develop an efficient iteration algorithm to implement our proposed MDP scheme. Finally, we validate and evaluate the performance of our proposed schemes through the numerical analyses, which show that our proposed schemes converge to a unit optimal solution.

Abstract: Edit paper A Peer-to-Peer (P2P) network is decentralised and relies on end-users capacities for a good delivery of service. The more end-users are sharing the content, the more powerful the system is. P2P applications are now widely used for many services. But P2P systems are still based on a end-to-end connectivity on the current IP-based Internet. For few years, a new networking paradigm emerged, the Content-Centric Networking (CCN) approach. In CCN, the user just cares about the content, not about the node that provides it. It takes as input some features of the P2P systems, applying them in the network itself. However, current CCN system forwards messages based on a defined forwarding table, filled with the advertisements of content providers. It does not take into the consideration of the fact that end-users that requested the contents can cache them for a while and thus be the potential sources of content, as it is in a P2P system. In this paper, we propose to take into consideration of the caching capacities of end-users in the CCN design so as to improve the content delivery in a CCN network. This is achieved by the design of a content-aware dynamic downstream forwarding scheme, and an adaptation of the current way to fill the forwarding table. Our evaluation shows that exploiting the end-users P2P caching facilities allows to improve the efficiency to the CCN network by reducing the memory requirements of the CCN node and by improving the data delivery to end-users via a closer delivery.