On the architecture and the design of P2P live streaming system schedulers

In this paper we analyze P2P live streaming systems. Through this analysis we obtain the crucial parameters for their performance in terms of bandwidth utilization, set-up time, fairness and stability. We propose a sender driven multi objective decision function for neighbor selection in order to adapt the distribution of available bandwidth to the overlay connections while simultaneously we further exploit the locality properties of an overlay. At last we develop and apply a receiver driven block selection with a content diffusion optimization algorithm that achieves fast and efficient diffusion of every block. The evaluation of our system reveals its very high levels of performance in terms of setup time, bandwidth utilization, its fair behavior in the distribution of available aggregate bandwidth in various nodes and the minimization of duplicate block transmissions. Finally by comparing our system with other recently developed we observe that it vastly outperforms.

[1]  Lifeng Sun,et al.  Understanding the Power of Pull-Based Streaming Protocol: Can We Do Better? , 2007, IEEE Journal on Selected Areas in Communications.

[2]  Venkata N. Padmanabhan,et al.  Analyzing and Improving BitTorrent Performance , 2005 .

[3]  Margo I. Seltzer,et al.  Supporting Network Coordinates on PlanetLab , 2005, WORLDS.

[4]  Miguel Castro,et al.  SplitStream: high-bandwidth multicast in cooperative environments , 2003, SOSP '03.

[5]  Keith W. Ross,et al.  Inferring Network-Wide Quality in P2P Live Streaming Systems , 2007, IEEE Journal on Selected Areas in Communications.

[6]  Nikolaos Efthymiopoulos,et al.  L-CAN: Locality Aware Structured Overlay for P2P Live Streaming , 2008 .

[7]  Reza Rejaie,et al.  PRIME: peer-to-peer receiver-driven mesh-based streaming , 2009, TNET.

[8]  Antony I. T. Rowstron,et al.  Pastry: Scalable, Decentralized Object Location, and Routing for Large-Scale Peer-to-Peer Systems , 2001, Middleware.

[9]  Keith W. Ross,et al.  A Measurement Study of a Large-Scale P2P IPTV System , 2007, IEEE Transactions on Multimedia.

[10]  Peter Druschel,et al.  Exploiting Network Proximity in Distributed Hash Tables , 2002 .

[11]  Reza Rejaie,et al.  Mesh or Multiple-Tree: A Comparative Study of Live P2P Streaming Approaches , 2007, IEEE INFOCOM 2007 - 26th IEEE International Conference on Computer Communications.

[12]  Venkata N. Padmanabhan,et al.  Analyzing and Improving a BitTorrent Networks Performance Mechanisms , 2006, Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications.

[13]  Laurent Massoulié,et al.  Randomized Decentralized Broadcasting Algorithms , 2007, IEEE INFOCOM 2007 - 26th IEEE International Conference on Computer Communications.

[14]  Emin Gün Sirer,et al.  Meridian: a lightweight network location service without virtual coordinates , 2005, SIGCOMM '05.

[15]  Rakesh Kumar,et al.  Stochastic Fluid Theory for P2P Streaming Systems , 2007, IEEE INFOCOM 2007 - 26th IEEE International Conference on Computer Communications.

[16]  Mark Handley,et al.  A scalable content-addressable network , 2001, SIGCOMM '01.

[17]  Peter Druschel,et al.  Pastry: Scalable, distributed object location and routing for large-scale peer-to- , 2001 .

[18]  Dimitri P. Bertsekas,et al.  Network optimization : continuous and discrete models , 1998 .