A novel cache optimization algorithm and protocol for video streaming in pure peer-to-peer networks

Authors:
Carlo Giulietti;Dan Schonfeld;Rashid Ansari
Affiliations:
University of Illinois at Chicago, Chicago, IL, USA;University of Illinois at Chicago, Chicago, IL, USA;University of Illinois at Chicago, Chicago, IL, USA
Venue:
Proceedings of the 2010 ACM workshop on Advanced video streaming techniques for peer-to-peer networks and social networking
Year:
2010

Citing 7
Cited 0

COPACC: An Architecture of Cooperative Proxy-Client Caching System for On-Demand Media Streaming

IEEE Transactions on Parallel and Distributed Systems
Cooperative Media Data Streaming with Scalable Video Coding

IEEE Transactions on Knowledge and Data Engineering
Improving Data Availability in P2P Streaming Systems Using Distributed Virtual Cache

ISM '08 Proceedings of the 2008 Tenth IEEE International Symposium on Multimedia
Peers' Segment Replacement Strategies in P2P Vod System Based On Client-Side Segmented Cache

CMC '09 Proceedings of the 2009 WRI International Conference on Communications and Mobile Computing - Volume 03
VOVO: VCR-Oriented Video-on-Demand in Large-Scale Peer-to-Peer Networks

IEEE Transactions on Parallel and Distributed Systems
Peer assisted video streaming with supply-demand-based cache optimization

IEEE Transactions on Multimedia - Special section on communities and media computing
Optimizing segment caching for peer-to-peer on-demand streaming

ICME'09 Proceedings of the 2009 IEEE international conference on Multimedia and Expo

Quantified Score

Hi-index	0.00

Visualization

Abstract

P2P networks offer scalable, inexpensive, and robust solutions to video streaming that overcome the shortcomings of the traditional client-server architecture. However, they suffer from user churn and node failure, thus posing a formidable challenge to the performance of P2P networks. In this paper, a novel cache allocation strategy designed to achieve optimality while factoring in characteristics of peers (i.e. reliability) and videos (i.e. demand, supply) is proposed. A distributed architecture based on local peer neighborhoods is developed, which provides an algorithm and protocol for peer caching, streaming, and routing. The allocation algorithm is obtained as a solution of a linear programming problem designed to maximize a utility function. A protocol based on local neighborhood relationships is proposed and handles data gathering, streaming, and peer communication. The solution results in a balanced workload distribution among peers and improved video availability. The performance of the proposed P2P allocation and streaming protocol is assessed through simulations and is shown to provide a significant improvement compared with existing methods.