Enabling QoS in peer to peer live streaming through dynamic bandwidth or playback rate control

Abstract

Live video streaming over internet is becoming popular more and more. Current commercial live video streaming systems are based either on a typical client-server (cloud) or on a peer-to-peer (P2P) architecture. The former is preferred for stability and QoS, provided that the system is not stretched beyond the server's bandwidth capacity, while the latter is scalable with small bandwidth and management cost. In this paper, we stabilize live streaming with a cloud assisted P2P live streaming system without sacrificing scalability. In order to achieve this we develop: i) a scalable gossip protocol with low bandwidth overhead for the cloud that monitors dynamically the total available bandwidth resources of the participating peers, ii) a control strategy that dynamically allocates from the cloud the exact amount of required bandwidth in order to guarantee the complete and on time stream distribution, iii) an alternative control strategy that dynamically adapts the playback rate to the available bandwidth resources. All these are achieved by modeling analytically through difference equations the dynamic relationship between the total bandwidth surplus (idle peers' bandwidth) and the playback rate. Finally, we evaluate our proposed architecture in a packet level simulator of a complete P2P live streaming system that designed in OPNET Modeler. Our evaluation testifies the uninterrupted and complete stream delivery even in adverse bandwidth changes.