FOCS '02 Proceedings of the 43rd Symposium on Foundations of Computer Science
FOCS '00 Proceedings of the 41st Annual Symposium on Foundations of Computer Science
Information Theory, Inference & Learning Algorithms
Information Theory, Inference & Learning Algorithms
Informed content delivery across adaptive overlay networks
IEEE/ACM Transactions on Networking (TON)
IEEE/ACM Transactions on Networking (TON) - Special issue on networking and information theory
OSDI '06 Proceedings of the 7th USENIX Symposium on Operating Systems Design and Implementation - Volume 7
ACM Transactions on Computer Systems (TOCS)
Rescuing Tit-for-Tat with Source Coding
P2P '07 Proceedings of the Seventh IEEE International Conference on Peer-to-Peer Computing
Epidemic live streaming: optimal performance trade-offs
SIGMETRICS '08 Proceedings of the 2008 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
Is There a Future for Mesh-Based live Video Streaming?
P2P '08 Proceedings of the 2008 Eighth International Conference on Peer-to-Peer Computing
T-Man: gossip-based overlay topology management
ESOA'05 Proceedings of the Third international conference on Engineering Self-Organising Systems
IEEE Transactions on Information Theory
A digital fountain approach to asynchronous reliable multicast
IEEE Journal on Selected Areas in Communications
A framework for secure and private P2P publish/subscribe
SSS'10 Proceedings of the 12th international conference on Stabilization, safety, and security of distributed systems
| Hi-index | 0.00 |
Gossip protocols are well known to provide reliable and robust dissemination protocols in highly dynamic systems. Yet, they suffer from high redundancy in the last phase of the dissemination. In this paper, we combine fountain codes (rateless erasure-correcting codes) together with gossip protocols for a robust and fast content dissemination in large-scale dynamic systems. The use of fountain enables to eliminate the unnecessary redundancy of gossip protocols. We propose the design of FoG , which fully exploits the first exponential growth phase (where the data is disseminated exponentially fast) of gossip protocols while avoiding the need for the shrinking phase by using fountain codes. FoG voluntarily increases the number of disseminations but limits those disseminations to the exponential growth phase. In addition, FoG creates a split-graph overlay that splits the peers between encoders and forwarders. Forwarder peers become encoders as soon as they have received the whole content. In order to benefit even further and quicker from encoders, FoG biases the dissemination towards the most advanced peers to make them complete earlier. We assess FoG through simulation. We show that FoG outperforms by 50% a simple push protocol with respect to overhead and improves by 30% the termination time.