iSync: A High Performance and Scalable Data Synchronization Protocol for Named Data Networking
by Wenliang Fu, Hila Ben Abraham, and Patrick Crowley
Extended abstract for poster. ACM Conference on Information-Centric Networking (ICN), September 2014.
This paper presents a high performance synchronization protocol for named data networking (NDN). The protocol, called iSync, uses a two-level invertible Bloom Filter (IBF) structure to support efficient data reconciliation. Multiple differences can be found by subtracting a remote IBF from a local IBF, and therefore, from a single round of data exchange. We evaluated iSync’s performance by comparing it to the default data synchronization protocol of CCNx. Experiments show that iSync is significantly faster for different network sizes and topologies, and it requires less overhead for synchronizing various file sizes.