Dynamic Adaptive Streaming over HTTP (DASH) is a recently proposed standardthat offers different versions of the same media content to adapt the deliveryprocess over the Internet to dynamic bandwidth fluctuations and different userdevice capabilities. The peer-to-peer (P2P) paradigm for video streaming allowsto leverage the cooperation among peers, guaranteeing to serve every videorequest with increased scalability and reduced cost. We propose to combinethese two approaches in a P2P-DASH architecture, exploiting the potentiality ofboth. The new platform is made of several swarms, and a different DASHrepresentation is streamed within each of them; unlike client-server DASHarchitectures, where each client autonomously selects which version to downloadaccording to current network conditions and to its device resources, we putforth a new rate control strategy implemented at peer site to maintain a goodviewing quality to the local user and to simultaneously guarantee thesuccessful operation of the P2P swarms. The effectiveness of the solution isdemonstrated through simulation and it indicates that the P2P-DASH platform isable to warrant its users a very good performance, much more satisfying than ina conventional P2P environment where DASH is not employed. Through a comparisonwith a reference DASH system modeled via the Integer Linear Programming (ILP)approach, the new system is shown to outperform such reference architecture. Tofurther validate the proposal, both in terms of robustness and scalability,system behavior is investigated in the critical condition of a flash crowd,showing that the strong upsurge of new users can be successfully revealed andgradually accommodated.
展开▼