Massive distribution of real-time video content is one of the big challenges in the Internet. Nowadays, there are several proposals that approximate to this goal, but none of them provide a QoS (Quality of Service) comparable to the DVB (Digital Video Broadcasting). This is a direct consequence of the design of the Internet and an uneﬃcient use of the capacity of the network. The Internet was created to provide the so called best-eﬀort service that basically means that you can transmit data through the network but the transmission time is unknown a priori. That time depends on several factors such as network failures, the network load and, obviously, the amount of sent data. These two last factors are directly related to the capacity of the network, a term also refered as (network) bandwidth.
In order to provide an acceptable QoS in real-time streaming scenarios one of the most important requirements to fulﬁll is to have a lower bound of the network capacity. However, most of the current solutions fail to take advantage of this resource. One of the reasons is that IP multicast has not achieved the expected popularity. This forces content sources to replicate the same data for diﬀerent receivers, resulting in a linear growth of the transmmited data when the number of receivers increases. In this situation, P2P (Peer-to-Peer) overlays can improve the performance of the real-time streaming services: given that all peers manage (and thus can share) the same content, the trasmission requirements at the source are drammatically reduced.
This work introduces P2PSP (Peer-to-Peer Straightforward Protocol). P2PSP is a set of transmission and machine behavior rules that helps increasing the QoS of real-time streaming systems. Basically, the P2PSP mimics the IP multicast solution by taking advantage of the transmission capacity of the peers which is typically wasted in pure C/S (Client/Server) services. We would like to stress that, although the P2PSP can be an standalone solution for small scenarios, its performance can be increased in massive scenarios by taking advantage of both paradigms: the C/S model and the P2P model.