Scalable name-based data synchronization for named data networking

TLDR

In Named Data Networking, data synchronization is essential, yet existing protocols only support full‑data sync, limiting pub‑sub applications that need selective subscriptions. This work introduces PSync, a scalable name‑based synchronization protocol that supports various synchronization scenarios. PSync embeds producers’ namespace and consumers’ subscription information in message names, enabling a single producer state to serve all consumers; it has been implemented in the NDN codebase and used to build a prototype pub‑sub module for building management. Experiments demonstrate that PSync scales with increasing consumers, subscriptions, and data streams, outperforming the state‑of‑the‑art Sync protocol in full‑data synchronization performance.

Abstract

In Named Data Networking (NDN), data synchronization plays an important role similar to transport protocols in IP. Many distributed applications, including pub-sub applications such as news and weather services, require a synchronization protocol where each consumer can subscribe to a different subset of a producer's data streams. However, existing Sync protocols support only full-data synchronization, which is a special case of this problem. We propose PSync to efficiently address different types of data synchronization. Names are used in PSync messages to carry producers' latest namespace information and each consumer's subscription information, which allows producers to maintain a single state for all consumers and enables consumers to synchronize with any producer that replicates the same data. We have implemented PSync in the NDN codebase and used it to develop a prototype pub-sub module for building management. Our experimental results show that PSync scales well as the number of consumers, subscriptions, and data streams increases and it outperforms the state-of-the-art Sync protocol in supporting full-data synchronization.