GroupSync: Scalable Distributed Dataset Synchronization for Named Data Networking
Yang Liu, Junfeng Ma, Hui Li, Feng Wang, Fuli Jiang · 2021 3rd International Academic Exchange Conference on Science and Technology Innovation (IAECST) · 2021
Dataset synchronization is a basic communication requirement for distributed applications such as multi-user chat, file transfer, and multi-user collaborative work. Named Data Networking (NDN) is a content-centric network architecture that forwards the content based on the name of the content. NDN is an important evolutionary direction for future networks. Various protocols have been developed in NDN to support dataset synchronization of distributed applications, such as ChronoSync, PSync, and VectorSync. These protocols generally have the problem of limited application scenarios and cannot adapt to large-scale deployment of the network. In this paper, we propose a scalable NDN distributed dataset synchronization protocol called GroupSync. GroupSync uses group synchronization to reduce network traffic overhead, which is scalable for the application of NDN in mobile scenarios. Our experimental results based on ndnSIM simulations show that GroupSync outperforms other protocols regarding dataset synchronization delay in the lossy mobility network and significantly reduces network traffic overhead.