TITLE:
P2P Traffic Optimization Based on Dynamic Network Information Aware
AUTHORS:
M. L. Lu, Z. Zhang, Y. Shi
KEYWORDS:
Peer-to-Peer, Traffic Optimization, Network Map, Link Cost Map, Traffic Lo-calization
JOURNAL NAME:
International Journal of Communications, Network and System Sciences,
Vol.10 No.5B,
May
26,
2017
ABSTRACT:
Peer-to-Peer (P2P) service may damage the interests of Internet Service Provider (ISP) because P2P traffic usually takes a lot of network link bandwidth and even overwhelms some network links. Aimed at the problem, mainstream solutions are usually optimizing P2P traffic through the interaction between applications and underlying network. However, current solutions still have two aspects of defects: one is that the interacted underlying network status information is immutable and can’t reflect the real-time dynamic changes because it is usually configured by ISP. The other is that some solutions may cause excessive traffic localization, which may greatly influence other services in the local network. In order to improve the above two defects and provide P2P users with better service experience, we propose an enhanced application layer traffic optimization scheme, in which more valuable network status information of underlying network is dynamically calculated and provided to P2P application. Extensive simulations demonstrate that our P2P traffic optimization scheme is superior to other solutions in terms of available bandwidth, resource transmission delay and user service experience.